Method and apparatus for determining renewable energy carrying capacity based on approximate dynamic programming

Abstract

The application relates to a renewable energy carrying capacity determination method and device based on approximate dynamic programming, a computer device, a storage medium and a computer program product. The method comprises the following steps: in response to a prediction request for the renewable energy carrying capacity of a target power system, acquiring constraint condition setting information of the prediction request input; constructing a power system planning problem for the target power system according to the constraint condition setting information; performing a solving operation on the power system planning problem by using approximate dynamic programming to obtain carrying capacity information for the target power system; and displaying display data generated based on the carrying capacity information. By using the method, the renewable energy carrying capacity of the power system can be efficiently calculated under the condition of considering various uncertainties of the power system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of renewable energy, in particular to a renewable energy carrying capacity determination method and device based on approximate dynamic programming, a computer device, a storage medium and a computer program product. BACKGROUND

[0002] The effective carrying capacity in the power system represents the additional load that can be provided in the power system due to the installation of new generation units under the same reliability level, which is the value established for a long time by traditional power plants. In recent years, wind power and solar power also have similar effective carrying capacity values, and the calculation of effective carrying capacity has become an important part of the integration of renewable resources in power system planning.

[0003] At present, in order to deal with a series of different forms of uncertainty that may occur in the power system, from hourly wind energy, solar energy and demand to daily and monthly changes in conventional unit fuel supply, the existing stochastic programming framework is usually used to deal with random conditions in optimization problems, and the uncertain information is usually represented as a scenario tree. However, the size of the scenario tree increases exponentially with the increase of the considered duration, and when solving practical application problems in the real world, it is necessary to solve large-scale models, which makes it impossible to efficiently calculate the renewable energy carrying capacity of the power system under consideration of various uncertainties of the power system.

[0004] Therefore, the conventional technology has the problem that the renewable energy carrying capacity of the power system cannot be efficiently calculated under consideration of various uncertainties of the power system. SUMMARY

[0005] Therefore, it is necessary to provide a renewable energy carrying capacity determination method and device based on approximate dynamic programming, a computer device, a computer readable storage medium and a computer program product, which can efficiently calculate the renewable energy carrying capacity of the power system under consideration of various uncertainties of the power system.

[0006] A renewable energy carrying capacity determination method based on approximate dynamic programming, the method comprising:

[0007] In response to a prediction request for the renewable energy carrying capacity of a target power system, acquiring constraint condition setting information input by the prediction request;

[0008] According to the constraint condition setting information, constructing a power system planning problem for the target power system; wherein the operation cost of the target power system meets the preset operation cost as the solution target of the power system planning problem, and the load expected loss constraint condition and the random safety constraint condition as the constraint condition of the power system planning problem;

[0009] The power system planning problem is solved by using approximate dynamic programming to obtain the carrying capacity information of the target power system. The carrying capacity evaluation information represents the renewable energy carrying capacity of the target power system.

[0010] Display data generated based on the carrying capacity information is displayed.

[0011] In one embodiment, the solution target is composed of a first solution target and a second solution target. The first solution target includes solution targets corresponding to different operating states of the day. The second solution target includes solution targets corresponding to load shedding of the day and solution targets corresponding to subsequent load shedding.

[0012] The first solution target is represented as:

[0013] ;

[0014] wherein, is a state variable representing the current state of the target power system; represents a value function about the state variable ; represents the value under the initial state ; E represents the expected value of the value under uncertainty; represents an uncertain parameter or a random variable;

[0015] The second solution target is represented as:

[0016] ;

[0017] wherein, is a probability function representing the probability under scenario ; represents the rated capacity of the device ; represents the current of the device under scenario , time , and time period ; represents the operating cost of the device under mode ; represents the power flow of the device under scenario , time , time period , and mode ; and represent scenarios Below, equipment In time Time period Start-up and downtime costs; Representation of context Next time Time period Load shedding amount on busbar b; The load shedding penalty coefficient represents the economic loss from load shedding.

[0018] In one embodiment, the constraints include a first constraint, a second constraint, a third constraint, a fourth constraint, and a fifth constraint.

[0019] The first constraint is expressed as:

[0020] ;

[0021] in, Characterization in the The optimal value function of stage corresponds to stage 1. The optimal value obtained after minimizing the cost over all time periods; Characterization in stages ,time and scenarios Below, equipment The startup status; Representation in pattern Lower device Operating costs; Characterization in stages ,time and scenarios Below, equipment In mode The power generation or load under the current conditions; Characterization in stages and time Below, equipment Startup costs; Representation stage and time Below, equipment Downtime costs; The load shedding penalty coefficient characterizes the economic loss incurred by load shedding. Characterization in stages ,time and scenarios Below, the load shedding amount on busbar b; Characterization in stages ,time and scenario n-1, the device of coupling coefficient; characterizing the nominal capacity of the device ; characterizing the estimated start-up state of the device , at stage , at time and scenario ; characterizing the stage , at time and scenario , the estimated power production or load of the device in mode ; characterizing the average load shedding penalty coefficient; characterizing the load forecast of all units at stage , at time and scenario ;

[0022] The second constraint is expressed as:

[0023] ;

[0024] wherein, characterizes the power flow of the i unit at time t and scenario n; characterizes the power flow forecast of the i unit on bus m at time t and scenario n;

[0025] The third constraint is expressed as:

[0026] ;

[0027] wherein, characterizes the current of the i unit at time t and scenario n; characterizes the current forecast of the i unit at time t and scenario n;

[0028] The fourth constraint is:

[0029] ;

[0030] wherein, characterizes the minimum power production of the device ; characterizes the maximum power production of the device ;

[0031] The fifth constraint is:

[0032] ;

[0033] wherein, characterize load demand under phases , times and scenarios; characterize safety margin for representing additional power generation needed under load demand.

[0034] In one of the embodiments, the power system planning problem is solved by using approximate dynamic programming to obtain the carrying capacity information for the target power system, including:

[0035] obtaining initialization state data of the target power system;

[0036] based on approximate dynamic programming, iteratively updating the initialization state data to obtain an iteratively updated result;

[0037] in the case where the iteratively updated result meets a preset convergence condition, determining the carrying capacity information of the target power system based on the iteratively updated result.

[0038] In one of the embodiments, the initialization state data is iteratively updated based on approximate dynamic programming to obtain an iteratively updated result, including:

[0039] processing the initialization state data by using an approximate value function to obtain cost information of an expected remaining number of days; the expected remaining number of days is the remaining number of days in which the target power system cannot meet the load demand; the cost information is the operation cost of the target power system in the expected remaining number of days;

[0040] based on the cost information of the expected remaining number of days, updating the initialization state data to obtain updated state data;

[0041] updating the approximate value function based on the updated state data to obtain an updated approximate value function as the iteratively updated result.

[0042] In one of the embodiments, the operation cost includes load shedding cost, no-load cost, power generation cost, start-up cost, shut-down cost, and expected value cost of insufficient power within the dispatching range.

[0043] A renewable energy carrying capacity determination apparatus based on approximate dynamic programming, the apparatus comprising:

[0044] a response module configured to, in response to a prediction request for renewable energy carrying capacity of a target power system, obtain constraint condition setting information input by the prediction request;

[0045] ​The construction module is configured to construct a power system planning problem for the target power system according to the constraint condition setting information, wherein a preset operation cost of the operation cost of the target power system is taken as a solution target of the power system planning problem, and the load expected loss constraint condition and the random safety constraint condition are taken as constraint conditions of the power system planning problem;

[0046] The solution module is configured to solve the power system planning problem by using the approximate dynamic programming to obtain carrying capacity information of the target power system, wherein the carrying capacity evaluation information represents the renewable energy carrying capacity of the target power system.

[0047] The display module is configured to display display data generated based on the carrying capacity information.

[0048] A computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0049] A computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the above method.

[0050] A computer program product includes a computer program, and the computer program is executed by a processor to implement the steps of the above method.

[0051] The above-mentioned renewable energy carrying capacity determination method, device, computer device, storage medium and computer program product based on approximate dynamic programming, in response to a prediction request for the renewable energy carrying capacity of the target power system, obtain constraint condition setting information input by the prediction request, construct a power system planning problem for the target power system according to the constraint condition setting information, wherein a preset operation cost of the operation cost of the target power system is taken as a solution target of the power system planning problem, and the load expected loss constraint condition and the random safety constraint condition are taken as constraint conditions of the power system planning problem, the approximate dynamic programming is used to solve the power system planning problem to obtain carrying capacity information of the target power system, the carrying capacity evaluation information represents the renewable energy carrying capacity of the target power system, and display data generated based on the carrying capacity information is displayed. In this way, the variable cost under the consideration of uncertain factors can be minimized, the operation cost of the baseline situation with the predicted value can be minimized, the evaluation accuracy of the renewable energy effective carrying capacity under various uncertainties in the power system planning and the operation cost prediction accuracy of the power system can be effectively improved, compared with the traditional method, the dispatcher can make decisions according to the current information, and the calculation burden of predicting the future state is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the accompanying drawings needed to be used in the embodiments or the related art description will be briefly introduced. Obviously, the accompanying drawings in the following description only represent some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative effort based on these drawings.

[0053] Figure 1 An application environment diagram of a renewable energy carrying capacity determination method based on approximate dynamic programming in an embodiment;

[0054] Figure 2 A flowchart of a renewable energy carrying capacity determination method based on approximate dynamic programming in an embodiment;

[0055] Figure 3 A flowchart of a solution method of a stochastic security constrained unit commitment model based on approximate dynamic programming in an embodiment;

[0056] Figure 4 A flowchart of a renewable energy carrying capacity determination method based on approximate dynamic programming in another embodiment;

[0057] Figure 5 A structural block diagram of a renewable energy carrying capacity determination device based on approximate dynamic programming in an embodiment;

[0058] Figure 6 An internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION

[0059] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not intended to limit the present application.

[0060] The renewable energy carrying capacity determination method based on approximate dynamic programming provided by the embodiments of the present application can be applied to, for example Figure 1The application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data required by the server 104 to process. The data storage system can be integrated on the server 104, or placed on the cloud or other network servers. The server 104 acquires constraint condition setting information input by a prediction request for the renewable energy carrying capacity of the target power system in response to the prediction request; the server 104 constructs a power system planning problem for the target power system according to the constraint condition setting information; wherein the operation cost of the target power system meets the preset operation cost as the solving target of the power system planning problem, and the load expected loss constraint condition and the random safety constraint condition as the constraint condition of the power system planning problem; the server 104 uses approximate dynamic programming to solve the power system planning problem, and obtains the carrying capacity information of the target power system; the carrying capacity evaluation information represents the renewable energy carrying capacity of the target power system; the server 104 displays the display data generated based on the carrying capacity information. Among them, the terminal 102 can be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, Internet of Things devices and portable wearable devices, Internet of Things devices can be smart speakers, smart televisions, smart air conditioners, smart car devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server 104 can be realized by an independent server or a server cluster composed of multiple servers.

[0061] In one exemplary embodiment, as shown in Figure 2 , a renewable energy carrying capacity determination method based on approximate dynamic programming is provided, which is applied to the server 104 in Figure 1 for example, including the following steps 202 to 208. Among them:

[0062] Step 202, in response to a prediction request for the renewable energy carrying capacity of the target power system, acquire the constraint condition setting information input by the prediction request.

[0063] Among them, the target power system can be any power system in the field of energy management.

[0064] Among them, the prediction request can be a request for obtaining the renewable energy carrying capacity evaluation value of the target power system.

[0065] Among them, the constraint condition setting information can be the constraint condition set by the user for evaluating the renewable energy carrying capacity of the target power system.

[0066] Optionally, the server acquires the constraint condition set by the user in response to a prediction request initiated by the user for the renewable energy carrying capacity of the target power system.

[0067] At step 204, the power system planning problem for the target power system is constructed according to the constraint condition setting information; wherein the operation cost of the target power system meets the preset operation cost as the solution target of the power system planning problem, and the load expected loss constraint condition and the random safety constraint condition are taken as the constraint conditions of the power system planning problem.

[0068] The power system planning problem can refer to a planning task corresponding to energy planning of the target power system.

[0069] The preset operation cost can refer to the minimum operation cost.

[0070] The load expected loss constraint condition can refer to a constraint condition determined based on a load expected loss (LOLE) index, wherein the load expected loss (LOLE) index is an expectation that the available power supply cannot meet the hourly system load, which can also be expressed as the expected number of days that the target power system cannot meet the load demand.

[0071] The random safety constraint condition can refer to a constraint condition corresponding to a stochastic security constrained unit commitment (SCUC).

[0072] Optionally, the server constructs the power system planning problem for the target power system according to the constraint condition set by the user, wherein the minimum operation cost of the target power system is taken as the solution target of the power system planning problem, and the load expected loss constraint condition and the random safety constraint condition are taken as the constraint conditions of the power system planning problem.

[0073] At step 206, the power system planning problem is solved by using approximate dynamic programming to obtain the carrying capacity information of the target power system; and the carrying capacity evaluation information represents the renewable energy carrying capacity of the target power system.

[0074] Approximate Dynamic Programming (ADP) is a nonlinear optimization method that combines the ideas of Reinforcement Learning (RL) and Dynamic Programming (DP). Since the long-term SCUC formula is a large-scale, non-convex, non-deterministic polynomial time difficulty (NP-hard) problem with uncertain parameters, ADP decomposes the original problem into sub-problems to evaluate the effective carrying capacity of wind and solar energy while maintaining the same reliability level considering uncertain information. The one-year SCUC model can be solved iteratively by the following steps: Step 1: Start from the initial state of each iteration, iteration 0 represents the base case with predicted values; Step 2: Solve the daily deterministic optimization, use the value function approximation (VFA) to calculate the cost of the remaining days; Step 3: Update the value function approximation; Step 4: Obtain Monte Carlo samples of the cost and calculate the initial state of the second day; Step 5: Update the iteration and return to Step 1.

[0075] Optionally, the server uses approximate dynamic programming to solve the power system planning problem and obtains the carrying capacity value for the target power system.

[0076] Step 208, display the display data generated based on the carrying capacity information.

[0077] The display data can be a value represented by the carrying capacity information.

[0078] Optionally, the server displays the display data generated based on the carrying capacity information.

[0079] In the aforementioned method for determining renewable energy carrying capacity based on approximate dynamic programming, the constraint setting information of the prediction request input is obtained in response to a prediction request for the renewable energy carrying capacity of a target power system. Based on the constraint setting information, a power system planning problem for the target power system is constructed. The objective of the power system planning problem is to satisfy a preset operating cost for the operating cost of the target power system, and the constraints are load expectation loss constraints and stochastic security constraints. Approximate dynamic programming is used to solve the power system planning problem to obtain carrying capacity information for the target power system. The carrying capacity assessment information characterizes the renewable energy carrying capacity corresponding to the target power system. Display data generated based on the carrying capacity information is presented. Thus, while considering variable costs under uncertainties, the operating cost of the baseline scenario with predicted values ​​is minimized. This effectively improves the accuracy of assessing the effective carrying capacity of renewable energy under various uncertainties in power system planning and the accuracy of predicting the operating cost of the power system. Compared with traditional methods, this can assist dispatchers in making decisions based on current information and reduce the computational burden of predicting future states.

[0080] In another embodiment, the solution objective consists of a first solution objective and a second solution objective; the first solution objective includes solution objectives corresponding to different operating states on the same day; the second solution objective includes solution objectives corresponding to load reduction on the same day and solution objectives corresponding to subsequent load reduction situations.

[0081] Here, the first objective corresponds to a multi-stage stochastic linear programming problem, with stages 1, ..., d, ..., ND each day. The first objective can then be expressed as:

[0082] ;

[0083] In the above formula, These are state variables, representing the current state of the target power system; Characterizing about state variables The value function; Characterize the initial state The value under uncertainty; E represents the expected value under uncertainty; Characterizes uncertain parameters or random variables.

[0084] The second objective corresponds to each node in each stage. The minimum operating cost of each node is divided into a two-stage stochastic programming problem, including an independent solution for the uncertainty of the unloading time in the first stage and a scenario in the following days in the second stage. The second objective can then be expressed as:

[0085] ;

[0086] In the above formula, is a probability function representing the probability under the scenario ; represents the rated capacity of the device ; represents the current of the device under the scenario , time , time period ; represents the operating cost of the device under the mode ; represents the power flow of the device under the scenario , time , time period and mode ; and represent the start-up cost and shutdown cost of the device under the scenario at time , time period ; represents the load shedding amount on the bus b at time , time period under the scenario ; is a load shedding penalty coefficient representing the economic loss of load shedding.

[0087] The embodiment is to add the load shedding cost to the objective function of the stochastic SCUC problem, and the objective function of the stochastic SCUC problem is to minimize the operating cost of all generator units, generally including the no-load cost, the power generation cost, the start-up / shutdown cost and the expected value of insufficient power (EENS) cost in the entire scheduling range. The next embodiment is to add the load loss expectation (LOLE) index as a limit to the constraint set to represent the system reliability in the stochastic security constrained unit commitment (SCUC) on the basis of the embodiment. The existing stochastic security constrained unit commitment (SCUC) constraints include system power balance, actual power generation capacity limit of thermal power generator units, power generation limit of solar and wind power generator units, minimum start-up and shutdown time limit, system rotation and operating reserve requirement, unit climbing and descending limit and fuel and emission limit. The DC power transmission network safety constraints include node power balance, power transmission flow equation and power transmission flow limit. In addition, the present application also considers the daily operation coupling of each day, that is, the units The final commitment state and power dispatch are regarded as the initial state of units in the next day stochastic security-constrained unit commitment (SCUC) problem. In addition, the load shedding cost (also known as the value of lost load, VOLL) can be determined, which refers to the load shedding multiplied by the load shedding price to compensate for users, expressed in $ / kWh. In practical applications, there is an upper limit to the total load shedding power of each bus.

[0088] In another embodiment, the constraints include a first constraint, a second constraint, a third constraint, a fourth constraint, and a fifth constraint;

[0089] The first constraint is represented as:

[0090] ;

[0091] wherein, characterizes the optimal value function at stage corresponding to the optimal value obtained after cost minimization for all time periods in stage ; characterizes the start-up state of the device in stage , time , and scenario ; characterizes the operating cost of the device in mode ; characterizes the power generation or load of the device in mode in stage , time , and scenario ; characterizes the start-up cost of the device in stage and time ; characterizes the shutdown cost of the device in stage and time ; characterizes the load shedding penalty coefficient, which represents the economic loss of load shedding; characterizes the amount of load shedding on bus b in stage , time , and scenario ; characterizes the coupling coefficient of the device in stage , time , and scenario n-1; characterizes the rated capacity of the device ; characterizing the estimated start-up state of the device at a stage , a time and a scenario ; characterizing the estimated power production or load of the device at a stage , a time and a scenario in a mode ; characterizing the average load shed penalty factor; characterizing the load forecast of all units at a stage , a time and a scenario ;

[0092] The second constraint is expressed as:

[0093] ;

[0094] wherein characterizes the power flow of the i unit at time t and scenario n; characterizes the power flow forecast of the i unit on bus m at time t and scenario n;

[0095] The third constraint is expressed as:

[0096] ;

[0097] wherein characterizes the current of the i unit at time t and scenario n; characterizes the current forecast of the i unit at time t and scenario n;

[0098] The fourth constraint is:

[0099] ;

[0100] wherein characterizes the minimum power production of the device ; characterizes the maximum power production of the device ;

[0101] The fifth constraint is:

[0102] ;

[0103] wherein characterizes the load demand at a stage , a time and a scenario ; characterize the security margin, which represents the additional generation needed under the load demand.

[0104] The application takes state variables as the unit commitment and power dispatch , while decision variables represent the start-up index , shut-down index and load shedding (the amount of load shedding of bus b at time t). The stochastic security constrained unit commitment (SCUC) problem based on the value function approximation (VFA) can be decomposed into terms in the first constraint, the first term of the first constraint corresponds to the actual cost of the day, and the second term of the first constraint corresponds to the predicted cost and based on the state variables of the day (i.e. ). In practical applications, the predicted cost can be described by future operating variables, which include future power dispatch and future unit commitment state.

[0105] The application establishes the relationship between the predicted cost and the state variables of the day (i.e. and ), which corresponds to the second constraint and the third constraint.

[0106] The application maintains the consistency of the future power dispatch and the future unit commitment state, which corresponds to the fourth constraint, which is similar to the constraint of the traditional stochastic security constrained unit commitment (SCUC) problem.

[0107] The application considers future load shedding to ensure that the sub-problems based on approximate dynamic programming are equivalent to the original problem. The load shedding cost in the previous iteration is used to approximate the future load shedding cost in the current iteration. is a predefined threshold value used to indicate whether there is a predicted load shedding cost, which corresponds to the fifth constraint.

[0108] In another embodiment, the approximate dynamic programming is used to solve the power system planning problem to obtain the carrying capacity information of the target power system, including: obtaining the initialization state data of the target power system; based on the approximate dynamic programming, the initialization state data is iteratively updated to obtain an iterative update result; in the case that the iterative update result satisfies a preset convergence condition, the carrying capacity information of the target power system is determined based on the iterative update result.

[0109] Wherein, the initialization state data can be the initial state data of the current of the target power system, power, load, etc.

[0110] Optionally, the server obtains initialization state data of the target power system, iteratively updates the initialization state data based on the approximate dynamic programming, obtains an iterative update result, and determines the carrying capacity information of the target power system based on the iterative update result in a case where the iterative update result meets a preset convergence condition.

[0111] In this embodiment, the initialization state data of the target power system is obtained, the initialization state data is iteratively updated based on the approximate dynamic programming to obtain an iterative update result, and the carrying capacity information of the target power system is determined based on the iterative update result in a case where the iterative update result meets a preset convergence condition, so that the objective function can be solved by using the approximate dynamic programming, and the carrying capacity information of the target power system can be accurately determined.

[0112] In another embodiment, iteratively updating the initialization state data based on the approximate dynamic programming to obtain an iterative update result includes: processing the initialization state data by using an approximate value function to obtain cost information of an expected remaining number of days; the expected remaining number of days is a remaining number of days in which the target power system cannot meet load demand; the cost information is an operation cost of the target power system in the expected remaining number of days; updating the initialization state data based on the cost information of the expected remaining number of days to obtain updated state data; and updating the approximate value function based on the updated state data to obtain an updated approximate value function as the iterative update result.

[0113] Optionally, the server processes the initialization state data by using an approximate value function to obtain cost information of an expected remaining number of days, updates the initialization state data based on the cost information of the expected remaining number of days to obtain updated state data, and updates the approximate value function based on the updated state data to obtain an updated approximate value function as the iterative update result.

[0114] In this embodiment, the initialization state data is processed by using an approximate value function to obtain cost information of an expected remaining number of days; the expected remaining number of days is a remaining number of days in which the target power system cannot meet load demand; the cost information is an operation cost of the target power system in the expected remaining number of days; the initialization state data is updated based on the cost information of the expected remaining number of days to obtain updated state data; and the approximate value function is updated based on the updated state data to obtain an updated approximate value function as the iterative update result, so that optimization of the approximate value function can be implemented.

[0115] The key of the stochastic security constrained unit commitment model based on approximate dynamic programming is to constantly update the approximate value function, which can be adjusted adaptively to the unit commitment state and power dispatch of each day by learning the history of the previous days and the prediction information of the next days. The approximate value function is pushed forward in time, without enumerating all states in the state space. Therefore, compared with the classic backward dynamic programming, the computational burden can be reduced. Since the approximate value function contains integer variables, the derivative of the approximate value function is replaced by the difference in this patent. In addition, the variable can take the minimum value or the maximum value, and in each iteration, the slope will be updated using the standard update equation, while the initial slope is defined by the ratio of future load to current daily load. In addition, when the load shedding and no load shedding occur, the threshold is also updated between the current iteration and the previous iteration. Here, the step between iterations and is a decreasing function to maintain and ensure that the value of the slope will converge to a normal level. The flowchart of the whole method is shown in Figure 3 , wherein based on the initialization state of the target power system, the deterministic optimization of each day (corresponding to the first solving target) is solved, the cost of the remaining days is calculated using the approximate value function VFA; the value function approximation is updated; the Monte Carlo sample of the cost is obtained and the initial state of the second day is calculated; the approximate value function VFA is updated; it is judged whether to converge, in the case of determining convergence, the task is ended, and in the case of not converging, the initialization state is updated, and the step of solving the deterministic optimization of each day (corresponding to the first solving target) is returned to calculate the cost of the remaining days using the approximate value function VFA, until convergence.

[0116] In another embodiment, the operating cost includes load shedding cost, no-load cost, power generation cost, start-up cost, shutdown cost, and power shortage expectation cost within the dispatching range.

[0117] In another embodiment, as shown in Figure 4 , a renewable energy carrying capacity determination method based on approximate dynamic programming is provided, which is applied to the server 104 in Figure 1 for example, including the following steps:

[0118] Step S402, in response to a prediction request for the renewable energy carrying capacity of the target power system, obtain the constraint condition setting information input by the prediction request.

[0119] Step S404, according to the constraint condition setting information, construct a power system planning problem for the target power system; wherein the solving target of the power system planning problem is that the operating cost of the target power system meets the preset operating cost, and the constraint conditions of the power system planning problem are the load expectation loss constraint condition and the random security constraint condition.

[0120] Step S406, obtaining initialization state data of the target power system.

[0121] Step S408, based on approximate dynamic programming, iteratively updating the initialization state data to obtain an iteratively updated result.

[0122] Step S410, in a case where the iteratively updated result meets a preset convergence condition, determining, based on the iteratively updated result, carrying capacity information of the target power system; the carrying capacity evaluation information representing renewable energy carrying capacity corresponding to the target power system.

[0123] Step S412, displaying display data generated based on the carrying capacity information.

[0124] It should be noted that the specific limitations of the above steps can refer to the specific limitations of the above-mentioned method for determining renewable energy carrying capacity based on approximate dynamic programming.

[0125] It should be understood that, although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise stated herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or steps or stages in other steps.

[0126] Based on the same inventive concept, the embodiments of the present application also provide a renewable energy carrying capacity determination device based on approximate dynamic programming for implementing the above-mentioned renewable energy carrying capacity determination method based on approximate dynamic programming. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more renewable energy carrying capacity determination device embodiments based on approximate dynamic programming provided below can refer to the limitations of the renewable energy carrying capacity determination method based on approximate dynamic programming described above, and will not be repeated here.

[0127] In one exemplary embodiment, as shown in Figure 5 a renewable energy carrying capacity determination device based on approximate dynamic programming is provided, which includes a response module 502, a construction module 504, a solving module 506 and a display module 508, wherein:

[0128] The response module 502 is used to obtain the constraint setting information of the prediction request input in response to the prediction request for the renewable energy carrying capacity of the target power system.

[0129] Module 504 is used to construct a power system planning problem for the target power system based on the constraint setting information. The objective of the power system planning problem is to satisfy the preset operating cost of the target power system. The constraint conditions of expected load loss and stochastic security are used as the constraints of the power system planning problem.

[0130] Solver module 506 is used to solve the power system planning problem using approximate dynamic programming to obtain the carrying capacity information of the target power system; the carrying capacity assessment information characterizes the renewable energy carrying capacity of the target power system.

[0131] Display module 508 is used to display display data generated based on load-bearing capacity information.

[0132] In one embodiment, the solution objective consists of a first solution objective and a second solution objective; the first solution objective includes solution objectives corresponding to different operating states on the same day; the second solution objective includes solution objectives corresponding to load reduction on the same day and solution objectives corresponding to subsequent load reduction situations;

[0133] The first objective is expressed as:

[0134] ;

[0135] in, These are state variables, representing the current state of the target power system; Characterizing about state variables The value function; Characterize the initial state The value under uncertainty; E represents the expected value under uncertainty; Characterizes uncertain parameters or random variables;

[0136] The second objective is expressed as:

[0137] ;

[0138] in, Let be a probability function, representing the probability in the scenario. The probability of it happening; Characterization equipment Rated capacity; In the context ,time time period the device ; characterizing the device ; the operating cost of the device in the mode ; characterizing the scenario , time , time period and mode of the device ; and characterizing the scenario , time , time period of the start-up cost and shutdown cost of the device ; characterizing the scenario , time , time period of the load shedding amount on the bus b

[0139] is a load shedding penalty coefficient, characterizing the economic loss of load shedding.

[0140] In one of the embodiments, the constraint conditions include a first constraint condition, a second constraint condition, a third constraint condition, a fourth constraint condition and a fifth constraint condition;

[0141] ;

[0142] wherein, characterizing the optimal value function in the first stage, corresponding to the optimal value obtained after cost minimization for all time periods in the stage ; characterizing the start-up state of the device , time and scenario in the stage ; characterizing the operating cost of the device in the mode ; characterizing the power generation or load of the device , time and scenario in the stage in the mode ; characterizing the start-up cost of the device in the stage and time ; characterizing the stage and time The equipment down cost; characterizes the load shedding penalty coefficient, which represents the economic loss of load shedding; characterizes the load shedding amount on bus b at stage , time and scenario ; characterizes the coupling coefficient of equipment at stage , time and scenario n-1; characterizes the rated capacity of equipment ; characterizes the estimated start-up state of equipment at stage , time and scenario ; characterizes the estimated power generation or load of equipment at stage , time and scenario in mode ; characterizes the average load shedding penalty coefficient; characterizes the load forecast of all units at stage , time and scenario ;

[0143] The second constraint condition is represented as:

[0144] ;

[0145] wherein, characterizes the power flow of i unit at time t and scenario n; characterizes the power flow forecast of i unit on bus m at time t and scenario n;

[0146] The third constraint condition is represented as:

[0147] ;

[0148] wherein, characterizes the current of i unit at time t and scenario n; characterizes the current forecast of i unit at time t and scenario n;

[0149] The fourth constraint condition is:

[0150] ;

[0151] wherein, characterizing the minimum power generation of the device ; characterizing the maximum power generation of the device ;

[0152] The fifth constraint condition is:

[0153] ;

[0154] wherein, characterizing the load demand at the stage , time and scenario ; characterizing the safety margin, for indicating the additional power generation required under the load demand.

[0155] In one of the embodiments, the solving module 506 is specifically configured to acquire initialization state data of the target power system; perform iterative updating on the initialization state data based on the approximate dynamic programming, to obtain an iterative updating result; and in a case where the iterative updating result meets a preset convergence condition, determine the carrying capacity information of the target power system based on the iterative updating result.

[0156] In one of the embodiments, the solving module 506 is specifically configured to process the initialization state data by using the approximate value function, to obtain cost information of an expected remaining number of days; the expected remaining number of days is a remaining number of days in which the target power system cannot meet the load demand; the cost information is an operation cost of the target power system in the expected remaining number of days; update the initialization state data based on the cost information of the expected remaining number of days, to obtain updated state data; and update the approximate value function based on the updated state data, to obtain an updated approximate value function as the iterative updating result.

[0157] In one of the embodiments, the operation cost includes a load shedding cost, an idle load cost, a power generation cost, a start-up cost, a shut-down cost, and a cost of an expected value of insufficient power within a dispatching range.

[0158] The modules in the above-described renewable energy carrying capacity determination apparatus based on approximate dynamic programming can be all or partially implemented by software, hardware, and combinations thereof. The modules can be embedded in or independent of a processor in a computer device in a hardware form, or stored in a memory in the computer device in a software form, so as to be called and executed by a processor to perform operations corresponding to the modules.

[0159] In one exemplary embodiment, a computer device, which can be a server, is provided, and an internal structure diagram of the computer device can be as shown in Figure 6As shown in the figure. The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through the system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capability. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the renewable energy carrying capacity determination data based on approximate dynamic programming. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with the terminal outside through the network connection. The computer program is executed by the processor to realize a renewable energy carrying capacity determination method based on approximate dynamic programming.

[0160] Those skilled in the art can understand that, Figure 6 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0161] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the computer program being executed by the processor to make the processor execute the steps of the renewable energy carrying capacity determination method based on approximate dynamic programming. The steps of the renewable energy carrying capacity determination method based on approximate dynamic programming can be the steps of the renewable energy carrying capacity determination method based on approximate dynamic programming in each of the above embodiments.

[0162] In one embodiment, a computer readable storage medium is provided, storing a computer program, the computer program being executed by the processor to make the processor execute the steps of the renewable energy carrying capacity determination method based on approximate dynamic programming. The steps of the renewable energy carrying capacity determination method based on approximate dynamic programming can be the steps of the renewable energy carrying capacity determination method based on approximate dynamic programming in each of the above embodiments.

[0163] In one embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, causes the processor to perform the steps of the above-described method for determining renewable energy carrying capacity based on approximate dynamic programming. Here, the steps of the method for determining renewable energy carrying capacity based on approximate dynamic programming can be the steps of any of the above-described embodiments of the method for determining renewable energy carrying capacity based on approximate dynamic programming.

[0164] It can be understood by those skilled in the art that all or part of the processes in the above-described embodiments can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-described embodiments. Any reference to memory, database or other medium in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0165] Any combination of the technical features of the above embodiments can be made. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0166] The above-described embodiments are merely illustrative of several embodiments of the present application, which are described in more detail and in a specific manner, but should not be construed as limiting the scope of the patent of the present application. It should be noted that, for those of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method for determining a renewable energy carrying capacity based on approximate dynamic programming, the method comprising: The method comprises: ​ in response to a prediction request for the renewable energy carrying capacity of a target power system, obtaining constraint condition setting information input by the prediction request; according to the constraint condition setting information, constructing a power system planning problem for the target power system; wherein the operation cost of the target power system meets the preset operation cost as the solution target of the power system planning problem, and the load expected loss constraint condition and the random safety constraint condition as the constraint condition of the power system planning problem; solving the power system planning problem by using approximate dynamic programming to obtain the carrying capacity information of the target power system; the carrying capacity information represents the renewable energy carrying capacity corresponding to the target power system; comprising: obtaining the initialization state data of the target power system; based on approximate dynamic programming, the initialization state data is iteratively updated to obtain an iterative update result; in the case that the iterative update result meets the preset convergence condition, the carrying capacity information of the target power system is determined based on the iterative update result; specifically comprising: using the approximate value function to process the initialization state data to obtain the cost information of the expected remaining days; the expected remaining days are the remaining days that the target power system cannot meet the load demand; the cost information is the operation cost of the target power system in the expected remaining days; based on the cost information of the expected remaining days, the initialization state data is updated to obtain updated state data; based on the updated state data, the approximate value function is updated to obtain an updated approximate value function as the iterative update result; displaying display data generated based on the carrying capacity information.

2. The method of claim 1, wherein, The solution target is composed of a first solution target and a second solution target; the first solution target comprises a solution target corresponding to different operation states of the day; the second solution target comprises a solution target corresponding to the load shedding condition of the day and a solution target corresponding to the subsequent load shedding condition; The first solution target is expressed as: ； wherein is a state variable representing a current state of the target power system; represents a value function with respect to the state variable ; V represents a value under an initial state ; E represents a value expectation under uncertainty ; and represents an uncertain parameter or a random variable; The second solution target is expressed as: ； in, Let be a probability function, representing the probability in the scenario. The probability of it happening; Characterization equipment Rated capacity; In the context ,time Time period Below, equipment The current; Characterization equipment In mode Operating costs below; Representation of context ,time Time period and pattern Lower device Power flow; and Representation of context Below, equipment In time Time period Start-up and downtime costs; Representation of context Next time Time period Load shedding amount on busbar b; The load shedding penalty coefficient represents the economic loss from load shedding.

3. The method of claim 1, wherein, The constraint condition comprises a first constraint condition, a second constraint condition, a third constraint condition, a fourth constraint condition and a fifth constraint condition; The first constraint condition is expressed as: ； in, Characterization in the The optimal value function of stage corresponds to stage 1. The optimal value obtained after minimizing the cost over all time periods; Characterization in stages ,time and scenarios Below, equipment The startup status; Representation in pattern Lower device Operating costs; Characterization in stages ,time and scenarios Below, equipment In mode The power generation or load under the current conditions; Characterization in stages and time Below, equipment Startup costs; Representation stage and time Below, equipment Downtime costs; The load shedding penalty coefficient characterizes the economic loss incurred by load shedding. Characterization in stages ,time and scenarios Below, the load shedding amount on busbar b; Characterization in stages ,time And in scenario n-1, the device The coupling coefficient; Characterization equipment Rated capacity; Characterization in stages ,time and scenarios Below, equipment The estimated startup state; Representation stage ,time and scenarios Below, equipment In mode The estimated power generation or load is as follows; Characterizing the average load shedding penalty coefficient; Characterization in stages ,time and scenarios Load forecast for all units under; The second constraint condition is expressed as: ； wherein, characterizing the power flow of the i unit at time t and scenario n; characterizing the power flow prediction of the i unit on bus m at time t and scenario n; The third constraint condition is expressed as: ； wherein, characterizing a current of the i cell at time t and scenario n; characterizing a current prediction of the i cell at time t and scenario n; The fourth constraint condition is: ； wherein, characterization device of the minimum power generation; characterization device of the maximum power generation; The fifth constraint condition is: ； wherein, characterizing the load demand at a stage , time and scenario; characterizing a safety margin, for indicating an additional amount of power generation needed at the load demand.​ 4. The method of claim 1, wherein, The operation cost comprises load shedding cost, no-load cost, power generation cost, start-up cost, shutdown cost and power shortage expectation value cost within the dispatching range.

5. An approximate dynamic programming-based renewable energy carrying capacity determination apparatus, characterized by, The device comprises: a response module configured to, in response to a prediction request for the renewable energy carrying capacity of a target power system, obtain constraint condition setting information input by the prediction request; The constructing module is configured to construct a power system planning problem for the target power system according to the constraint condition setting information, wherein a preset operation cost of the target power system is taken as a solution target of the power system planning problem, and a load expectation loss constraint condition and a random safety constraint condition are taken as constraint conditions of the power system planning problem; The solving module is configured to solve the power system planning problem by using an approximate dynamic programming to obtain bearing capacity information of the target power system, wherein the bearing capacity information represents a renewable energy bearing capacity of the target power system. The solving module is specifically configured to obtain initialization state data of the target power system, update the initialization state data by using the approximate dynamic programming to obtain an iteration update result, and determine the bearing capacity information of the target power system based on the iteration update result when the iteration update result meets a preset convergence condition. The solving module is specifically configured to process the initialization state data by using an approximate value function to obtain cost information of an expected remaining number of days, wherein the expected remaining number of days is a remaining number of days in which the target power system cannot meet a load demand, the cost information is an operation cost of the target power system in the expected remaining number of days, the initialization state data is updated based on the cost information of the expected remaining number of days to obtain updated state data, the approximate value function is updated based on the updated state data to obtain an updated approximate value function as the iteration update result. The display module is configured to display display data generated based on the bearing capacity information. 6.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-5. The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 4.

7. A computer readable storage medium having stored thereon a computer program, characterized in that The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 4.

8. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 4.

Images