Distributed photovoltaic output robustness optimization method and system, electronic equipment and medium

Abstract

The invention discloses a distributed photovoltaic output robustness optimization method and system, an electronic device and a medium, and the method achieves the construction of a target function of the overall operation cost of a micro-grid, reduces the conservative property of a decision result, and improves the adaptability of a model. Moreover, the influence of the flexible resources of the demand side and the demand response on the power dispatching is considered, and the flexibility of the power dispatching of the micro-grid is improved, so that the robustness and economical efficiency of the operation of the micro-grid can be considered. And meanwhile, the target function is optimized and solved through the improved rabbet search algorithm, so that the global search capability of the rabbet search algorithm is improved, and the falling into a local optimal solution is avoided. Moreover, the confidence interval is generated through the QR algorithm to verify the credibility of the optimal solution, the uncertainty of the photovoltaic output and the load demand is considered through the confidence interval, a reasonable fluctuation range is provided for the photovoltaic output and the load demand, and the conservative property of a decision result is further reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of microgrids, and in particular, to a method and system for robust optimization of distributed photovoltaic output, an electronic device, and a computer-readable storage medium. Background Art

[0002] As an important part of small-scale power generation and distribution networks, microgrids integrate renewable energy, energy storage devices, and loads, and are gradually changing the energy structure of power systems, showing great technical and economic potential. Although microgrids can significantly reduce power grid energy losses and improve system efficiency, their stable operation faces severe challenges of intermittency and uncertainty of renewable energy output. Among them, the access of distributed photovoltaic power stations increases the complexity of the distribution network, and the uncertainty of their output also poses a threat to the stable operation of the power system. For example, the output of distributed photovoltaic power stations is affected by various factors such as light intensity, ambient temperature, and cloud cover. These factors have great uncertainty, which will cause the output of photovoltaic power stations to fluctuate frequently, bringing challenges to the operation of the distribution network. Therefore, how to achieve the stable and economic operation of microgrids under the uncertainty of distributed photovoltaic output has always been a research hotspot in this field.

[0003] The robust optimization method has unique advantages in dealing with uncertainty. It introduces an uncertain set in the modeling and describes the fluctuation range of random variables in the form of intervals, so that precise probability distribution information is not required. At present, it has been applied in small-scale power systems because it can ensure the reliability of system operation while sacrificing only a small amount of operating cost. However, most of the uncertain sets of traditional robust optimization methods adopt box-type uncertain sets. Although their broad boundaries can cover various scenarios, the optimization results are often too conservative, reducing the economy of system operation, and do not consider the flexible resources on the dispatching power demand side, nor the impact of demand response, restricting the flexibility of dispatching and making it difficult to balance the robustness and economy of microgrid operation. Summary of the Invention

[0004] The present invention provides a method and system for robust optimization of distributed photovoltaic output, an electronic device, and a computer-readable storage medium, which can balance the robustness and economy of microgrid operation and improve the accuracy of solving the optimal solution.

[0005] According to one aspect of the present invention, a method for robust optimization of distributed photovoltaic output is provided, including the following:

[0006] Adopt a two-stage scheduling strategy and combine the flexible resource scheduling and demand response on the demand side to construct an objective function for the overall operating cost of the microgrid and set the constraint conditions of the objective function;

[0007] An improved transit search algorithm is used to optimize and solve the objective function to obtain the optimal solution;

[0008] The QR algorithm is used to predict and generate a confidence interval based on the optimal solution, and the credibility of the optimal solution is judged according to whether the optimal solution falls within the confidence interval.

[0009] Furthermore, the objective function of the overall operation cost of the microgrid is:

[0010]

[0011] Among them, Y one represents the cost target in the pre-scheduling stage, and Y two represents the cost target in the re-scheduling stage. and represent the start-up cost and shutdown cost of the microturbine MT respectively. and represent the start-up cost and shutdown cost of the distributed generation DG respectively. d MT 、d DG 、d BS and d grid represent the operating cost of the microturbine MT, the operating cost of the distributed generation DG, the charging and discharging cost of the energy storage system, and the power trading cost between the microgrid and the main grid respectively. U represents the uncertainty set, u represents the uncertainty variable, including the photovoltaic power generation and the load demand. x represents the decision variable set in the pre-scheduling stage, y represents the decision variable set in the re-scheduling stage, Q represents the total power reduction of the demand-side response, and q t represents the reduction power of the demand side per unit time. T represents the time length, and F(t) represents the price cost of calling the demand-side flexibility resources. E1 and E2 represent the unmet amounts of the valley filling demand response and peak shaving demand response of the demand-side load respectively. f1 and f2 represent the unit compensation costs of the valley filling demand response and peak shaving demand response respectively. DR represents the cost required to implement the demand-side response. N DR represents the number of users of the demand-side response. v represents the critical power value for triggering the compensation price. S DR represents the compensation amount above the critical power value. L DR represents the compensation amount below the critical power value. z a and z b represent the compensation ratio coefficients above and below the critical power value respectively.

[0012] Furthermore, the process of using the improved transit search algorithm to optimize and solve the objective function includes the following contents:

[0013] During the development stage of the transit search algorithm, a sine-cosine mutation strategy is adopted to update the current solution, and the update expression is:

[0014]

[0015] where f E represents the current solution obtained by the transit search algorithm during the development stage, f E ′ represents the updated solution, r1, r2, r3, and r4 are all random values, represents the position of the optimal solution in the j-th dimension after t iterations.

[0016] Furthermore, the magnitude of the random value r1 decreases as the number of iterations increases.

[0017] Furthermore, the process of optimizing and solving the objective function using the improved transit search algorithm includes the following:

[0018] During the neighborhood stage of the transit search algorithm, the characteristics of the Cauchy distribution are used to update the current solution, and the update expression is:

[0019]

[0020] where f p represents the current solution obtained by the transit search algorithm during the neighborhood stage, θ represents a random value drawn from the Cauchy distribution, and f p ′ represents the updated solution.

[0021] Furthermore, the process of using the QR algorithm to generate a confidence interval based on the optimal solution prediction and judging the credibility of the optimal solution according to whether the optimal solution falls within the confidence interval includes the following:

[0022] The QR algorithm is used to predict the estimated values at the 0.05 quantile and 0.95 quantile based on the optimal solution to generate a 90% confidence interval. If the optimal solution falls within the 90% confidence interval, the optimal solution is determined to be credible.

[0023] Furthermore, the constraint conditions of the objective function include the power supply-demand balance constraint condition, and its expression is:

[0024]

[0025] where represents the total actual power generation output of thermal power units, represents the sum of the actual power generation outputs of wind power units and photovoltaic power units, S W represents the actual power generation output of wind power units, S P represents the actual power generation output of photovoltaic power units, and respectively represent the total charging power and the total charging power of the energy storage node, represents the total demand power on the load side, S load 、S Z 、S f and S d respectively represent the reference load power, valley filling type demand response power, peak shaving type demand response power and power shortage load power.

[0026] In addition, the present invention also provides a robustness optimization system for distributed photovoltaic output, including:

[0027] A target function construction module, which is used to adopt a two-stage scheduling strategy and combine the flexible resource scheduling and demand response on the demand side to construct the target function of the overall operation cost of the microgrid and set the constraint conditions of the target function;

[0028] An optimization solution module, which is used to optimize and solve the target function by using an improved transit search algorithm to obtain the optimal solution;

[0029] A credibility judgment module, which is used to generate a confidence interval based on the optimal solution prediction by using the QR algorithm and judge the credibility of the optimal solution according to whether the optimal solution falls within the confidence interval.

[0030] In addition, the present invention also provides an electronic device, including a processor and a memory. A computer program is stored in the memory, and the processor is used to execute the steps of the method described above by calling the computer program stored in the memory.

[0031] In addition, the present invention also provides a computer-readable storage medium for storing a computer program for robustly optimizing distributed photovoltaic output. The computer program executes the steps of the method described above when running on a computer.

[0032] The present invention has the following beneficial effects:

[0033] The robust optimization method for distributed photovoltaic output of the present invention adopts a two-stage scheduling strategy and constructs an objective function for the overall operating cost of the microgrid by combining the flexible resource scheduling on the demand side and demand response. This not only reduces the conservatism of the decision-making result, enhances the self-adaptability of the model, but also takes into account the impact of flexible resources and demand response on the demand side on power scheduling, improving the flexibility of power scheduling in the microgrid, so as to balance the robustness and economy of the microgrid operation. At the same time, the objective function is optimized and solved by an improved transiting search algorithm, which improves the global search ability of the transiting search algorithm, avoids falling into local optimal solutions, and improves the accuracy of solving the optimal solution. Moreover, instead of directly adopting the traditional box-type uncertainty set to limit the variation range of distributed photovoltaic output and load demand, a confidence interval is generated by the QR algorithm to verify the credibility of the optimal solution, and the uncertainty of photovoltaic output and load demand is considered through the confidence interval, providing a reasonable fluctuation range for it, which is conducive to further reducing the conservatism of the decision-making result.

[0034] In addition, the robust optimization system for distributed photovoltaic output of the present invention also has the above advantages.

[0035] In addition to the purposes, features and advantages described above, the present invention has other purposes, features and advantages. The present invention will be further described in detail below with reference to the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings forming a part of this application are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0037] Figure 1 is a schematic flowchart of the robust optimization method for distributed photovoltaic output of the preferred embodiment of this application;

[0038] Figure 2 is a schematic diagram of generating a confidence interval based on the optimal solution by the QR algorithm in the preferred embodiment of this application;

[0039] Figure 3 is a schematic diagram of the module structure of the robust optimization system for distributed photovoltaic output of another embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] It should be noted that, without conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the drawings and in combination with the embodiments.

[0041] Referring to Figure 1 , the preferred embodiment of this application provides a robust optimization method for distributed photovoltaic output, including the following content:

[0042] Step S1: Adopt a two-stage scheduling strategy and combine the flexible resource scheduling and demand response on the demand side to construct the objective function of the overall operating cost of the microgrid, and set the constraint conditions of the objective function.

[0043] Step S2: Use an improved transiting search algorithm to optimize and solve the objective function to obtain the optimal solution.

[0044] Step S3: Use the QR algorithm to predict and generate the confidence interval based on the optimal solution, and judge the credibility of the optimal solution according to whether the optimal solution falls within the confidence interval.

[0045] It can be understood that for the robust optimization method of distributed photovoltaic output in this embodiment, by adopting a two-stage scheduling strategy and combining the flexible resource scheduling and demand response on the demand side to construct the objective function of the overall operating cost of the microgrid, not only the conservatism of the decision-making result is reduced, the self-adaptability of the model is enhanced, but also the influence of the flexible resources and demand response on the demand side on power scheduling is considered, and the flexibility of the microgrid power scheduling is improved, so as to take into account the robustness and economy of the microgrid operation. At the same time, by using an improved transiting search algorithm to optimize and solve the objective function, the global search ability of the transiting search algorithm is improved, the local optimal solution is avoided, and the accuracy of the optimal solution is improved. Moreover, instead of directly using the traditional box-type uncertainty set to limit the change range of distributed photovoltaic output and load demand, the QR algorithm is used to generate the confidence interval to verify the credibility of the optimal solution, and the uncertainty of photovoltaic output and load demand is considered through the confidence interval to provide a reasonable fluctuation range, which is beneficial to further reducing the conservatism of the decision-making result.

[0046] Among them, in the step S1, first optimize the robustness of the distributed photovoltaic output and construct the following two-stage optimal scheduling model:

[0047]

[0048] Among them, Y one represents the cost target in the day-ahead pre-scheduling stage, Y two represents the cost target in the day-ahead re-scheduling stage, U represents the uncertainty set, u represents the uncertainty variable, including photovoltaic power generation and load demand, x represents the decision variable set in the pre-scheduling stage, y represents the decision variable set in the re-scheduling stage, and F(x, u) represents the feasible region formed by the decision of the decision variable set x in the pre-scheduling stage and the uncertainty variable u. C T X represents the total start-stop cost of the micro-turbine MT and distributed generation DG, D T Y represents the total of other costs in the day-ahead scheduling stage except the start-stop costs of MT and DG.

[0049] In the pre-scheduling stage, by clarifying the operating conditions of the micro-turbine MT and the distributed generator DG, it can be ensured that the microgrid can effectively adapt to different scenarios of photovoltaic power generation output and load demand. The core goal of the pre-scheduling stage is to optimize the system state. The cost objective of the day-ahead pre-scheduling stage can be expressed as: and represent the start-up cost and shutdown cost of the micro-turbine MT respectively, and represent the start-up cost and shutdown cost of the distributed generation DG respectively.

[0050] Based on the output results of the pre-scheduling stage, in the re-scheduling stage, on the premise that the operating state of the microgrid has been determined, a deterministic optimization strategy is used to calculate the output power of the equipment and the power transaction between the microgrid and the main grid, so as to ensure the stability and reliability of the microgrid operation. The objective function of the re-scheduling can be expressed as: where, d MT 、d DG 、d bS and d grid represent the operating cost of the micro-turbine MT, the operating cost of the distributed generation DG, the charge and discharge cost of the energy storage system, and the power transaction cost between the microgrid and the main grid respectively.

[0051] Then, constraint conditions are set for the two-stage optimal scheduling model, specifically including:

[0052] 1). Output power limit conditions of new energy power generation devices: where, p max represents the expected maximum power generation potential of the wind farm, Δp represents the load that can be cut by the wind farm, p represents the wind power generation that cannot be utilized by the wind farm, represents the expected maximum power generation potential of the photovoltaic power station, W pv represents the photovoltaic power generation that cannot be utilized by the photovoltaic power station, ΔW pv represents the load that can be cut by the photovoltaic power station.

[0053] 2). Capacity constraint conditions of energy storage devices: W = w1 + w2 + w3, where, W represents the capacity of the energy storage device, w1 represents the value of demand response, w2 represents the power output contributed by the energy storage device, and w3 represents the electricity required by the load. In addition, in order to ensure the long-term service life of the energy storage device and avoid its overcharging and over-discharging, it is necessary to limit the energy storage capacity within a specific range to manage the stored electricity, which can be expressed as: where, W min and W max represent the minimum and maximum values of the electricity of the energy storage device respectively, β represents the natural loss rate of the electricity of the energy storage device per hour, W tIndicates the power of the energy storage device at time t, W t-1 Indicates the power of the energy storage device at time t-1, ΔW t Indicates the value of the power change of the energy storage device at time t.

[0054] 3), Constraints of flexible load: Q min ≤Q 柔 ≤Q max , Q min and Q max respectively represent the lower limit power and the upper limit power of the flexible load.

[0055] Optionally, in order to ensure the stability of the microgrid operation, the present invention also adds a power supply-demand balance constraint condition in the constraint conditions, which can be expressed as:

[0056]

[0057] Among them, represents the total actual power generation output of the thermal power unit, represents the sum of the actual power generation outputs of the wind power unit and the photovoltaic power unit, S W represents the actual power generation output of the wind power unit, S P represents the actual power generation output of the photovoltaic power unit, and respectively represent the total charging power and the total discharging power of the energy storage node, represents the total demand power on the load side, S load , S Z , S f and S d respectively represent the reference load power, the valley-filling type demand response power, the peak-shaving type demand response power and the power shortage load power. The power shortage load power is the load amount that the microgrid fails to meet. It can be understood that at each moment of the microgrid operation, the above power supply-demand balance constraint conditions need to be satisfied to ensure the stable and reliable operation of the microgrid.

[0058] Next, considering the current power demand response scenario, the demand side can flexibly adopt demand response measures of valley filling or peak shaving according to the demand in different time periods. It is necessary to dig out the reduction potential of the photovoltaic output on the operation cost of the demand side. When the flexible resources in the demand side scenario are not enough to support the safe and economic operation of the microgrid, it is necessary to minimize the cost of calling the flexible resources on the demand side within the simulation period, which can be expressed as: Among them, F(t) represents the price cost of calling on demand-side flexibility resources. The demand-side flexibility resources include various electricity loads in the microgrid. E1 and E2 respectively represent the regulation amounts of valley filling demand response and peak shaving demand response of the demand-side load, and f1 and f2 respectively represent the unit compensation costs of valley filling demand response and peak shaving demand response. Moreover, in the case of power supply-demand imbalance, as a strategy, demand response enables users to adjust their electricity consumption behaviors according to electricity price fluctuations and load reward policies to assist microgrid dispatching and ensure the stable operation of the power grid in the short term. The cost required to implement demand response also needs to be considered and can be expressed as: Among them, DR represents the cost required to implement demand response, that is, the subsidy cost, N DR represents the number of users of demand response, v represents the critical power value for triggering the compensation price, S DR represents the compensation amount above the critical power value, L DR represents the compensation amount below the critical power value, z a and z b respectively represent the compensation ratio coefficients above and below the critical power value.

[0059] Therefore, the objective function of the overall operating cost of the microgrid can be constructed as:

[0060]

[0061] Among them, Q represents the total power reduction of demand response, q t represents the power reduction per unit time of the demand side, and T represents the time length.

[0062] In addition, in the step S2, an improved transit search algorithm is used to optimize the parameters of the above objective function to obtain the optimal solution. Among them, the traditional transit search algorithm simulates the natural process of celestial body operation. It uses photometric, distance, and transit observation data as key decision variables and simulates the optimization mechanism driven by natural phenomena in astronomy. It includes a total of five stages: the galaxy stage, the transit stage, the planet stage, the neighborhood stage, and the exploitation stage. Its solution process belongs to the prior art and will not be elaborated here. However, the traditional transit search algorithm strongly depends on the initial value and lacks global search ability, and it is very easy to fall into the trap of local optimum. Therefore, the present invention improves the existing transit search algorithm. Among them, the process of using the improved transit search algorithm to optimize and solve the objective function includes the following content:

[0063] In the exploitation stage of the transit search algorithm, a sine-cosine mutation strategy is adopted to update the current solution, and the update expression is:

[0064]

[0065] Among them, f E represents the current solution obtained by the transit search algorithm in the development stage, and f E ' represents the updated solution. r1, r2, r3, and r4 are all random values, representing the position of the optimal solution in the j-th dimension after t iterations.

[0066] It can be seen from the above formula that the update algorithm of the present invention includes four random numbers r1, r2, r3, and r4. r1 is used to control the search range of the algorithm. When r1 is large, the algorithm tends to perform global search and explore unknown regions. When r1 is small, the algorithm tends to perform local development and fully search the neighborhood of the current solution; r2 determines the search direction. r2 is randomly selected within the range of [0, 2π]. Due to the periodicity of the sine function and cosine function, different values of r2 will result in different search directions, which helps to explore uniformly in the solution space and avoid falling into local optimal solutions. The randomness of r2 increases the diversity of the search and enables search in different directions, increasing the possibility of finding the global optimal solution; r3 is used to control the influence degree of the current optimal solution on the search individual. r3 is randomly selected within [0, 1]. It determines the degree to which the search individual approaches the current optimal solution when updating the position. A larger value of r3 will make the search individual closer to the current optimal solution, while a smaller value of r3 will make the search individual search in a wider area; r4 can determine whether to use the sine function or cosine function to update the position of the search individual. r4 is randomly selected within the range of [0, 1]. When r4 ≤ 0.5, the sine function is used to update the position. When r4 > 0.5, the cosine function is used to update the position. The randomness of r4 can switch between the sine and cosine functions, thus increasing the diversity and flexibility of the search. Therefore, these random factors in the update algorithm of the present invention enable the current solution to have different movement methods and positions in each iteration, thus avoiding the algorithm from falling into local optimal solutions and reducing the sensitivity to the initial value. The parameters and search direction can be dynamically adjusted, so as to more effectively find the optimal solution, further reduce the dependence on the initial value, and improve the diversity and robustness of the update algorithm.

[0067] Optionally, the magnitude of the random value r1 decreases with the increase of the number of iterations. Wide global search can be performed in the initial stage of iteration to reduce the dependence on the initial value, and fine local development can be performed in the later stage of iteration to find the optimal solution. In addition, this decreasing method can be implemented by linear decrease, non-linear decrease, or other strategies to meet the requirements of different problems.

[0068] In addition, the process of optimizing and solving the objective function by using the improved transit search algorithm includes the following contents:

[0069] During the neighborhood stage of the transit search algorithm, the characteristics of the Cauchy distribution are used to update the current solution, and the update expression is:

[0070]

[0071] where f p represents the current solution obtained by the transit search algorithm in the neighborhood stage, that is, the location of the current planetary body, θ represents a random value drawn from the Cauchy distribution, and f p ′ represents the updated solution, that is, the changed position of the planetary body.

[0072] It can be understood that in the neighborhood stage of the traditional transit search algorithm, neighborhood search is performed on the surrounding area, and the average value of the planetary body in N runs is taken to determine the final position, which is likely to fall into the local optimal trap. However, the update algorithm of the present invention utilizes the mutation characteristics of the Cauchy distribution, draws random values from the Cauchy distribution to greatly mutate the current position of the planetary body obtained in the neighborhood stage, enhances the diversity in the search process, thereby optimizing the global search efficiency, and helps the update algorithm jump out of the current local optimal area and enter a new search space.

[0073] In addition, in the step S3, the process of generating a confidence interval based on the optimal solution prediction by using the QR algorithm and judging the credibility of the optimal solution according to whether the optimal solution falls within the confidence interval includes the following content:

[0074] Using the QR algorithm to obtain the estimated values at the 0.05 quantile and 0.95 quantile based on the optimal solution prediction to generate a 90% confidence interval. If the optimal solution falls within the 90% confidence interval, it is determined that the optimal solution is credible.

[0075] It can be understood that the QR (quantile regression) algorithm can explain the functional relationship between the explanatory variable and the explained variable at a specific quantile. After obtaining the optimal solution in step S2, the QR method can be used to estimate the estimated values at the 0.05 quantile and 0.95 quantile to generate a 90% confidence interval. If the optimal solution falls within the 90% confidence interval, it is determined that the optimal solution is credible. For example, as Figure 2 shown, assuming that the objective function of calculating Y one or Y two is the blue curve, then the QR algorithm is used to generate the red interval based on the blue curve. If the blue curve falls within the red interval, it is considered that the optimal solution has a 90% probability of being correct. If the blue curve falls outside the red interval, it is considered that there may be calculation errors or poor model errors, and it is determined that the optimal solution is incorrect.

[0076] In addition, in order to better verify the model effect, the MAE and MAPE can also be used to quantify the accuracy of the model. Among them, the calculation formulas of MAE and MAPE are:

[0077]

[0078] Among them, F(t) represents the actual measurement value at time t, while f(t) is the model prediction value at the same time. N is the total number of points in the time series. When the values of MAE and MAPE are smaller, it indicates that the optimization result of photovoltaic output is closer to the measured value, that is, the higher the accuracy of photovoltaic output optimization.

[0079] In addition, as Figure 3 shown, another embodiment of the present invention also provides a robust optimization system for distributed photovoltaic output, preferably adopting the robust optimization method for distributed photovoltaic output as described above, including:

[0080] An objective function construction module, which is used to adopt a two-stage scheduling strategy and combine the flexible resource scheduling and demand response on the demand side to construct the objective function of the overall operation cost of the microgrid and set the constraint conditions of the objective function;

[0081] An optimization and solution module, which is used to optimize and solve the objective function by using an improved transit search algorithm to obtain the optimal solution;

[0082] A credibility judgment module, which is used to generate a confidence interval based on the optimal solution by using the QR algorithm and judge the credibility of the optimal solution according to whether the optimal solution falls within the confidence interval.

[0083] It can be understood that the robust optimization system for distributed photovoltaic output in this embodiment adopts a two-stage scheduling strategy and combines the flexible resource scheduling and demand response on the demand side to construct the objective function of the overall operation cost of the microgrid. It not only reduces the conservatism of the decision-making result, enhances the self-adaptability of the model, but also considers the influence of the flexible resources and demand response on the demand side on power scheduling, improves the flexibility of the microgrid power scheduling, and thus can take into account the robustness and economy of the microgrid operation. At the same time, by using the improved transit search algorithm to optimize and solve the objective function, the global search ability of the transit search algorithm is improved, avoiding falling into the local optimal solution and improving the accuracy of solving the optimal solution. And instead of directly adopting the traditional box-type uncertainty set to limit the change range of distributed photovoltaic output and load demand, the credibility of the optimal solution is verified by generating a confidence interval through the QR algorithm, and the uncertainty of photovoltaic output and load demand is considered through the confidence interval, providing a reasonable fluctuation range for it, which is beneficial to further reducing the conservatism of the decision-making result.

[0084] In addition, another embodiment of the present invention also provides an electronic device, including a processor and a memory. A computer program is stored in the memory, and the processor is used to execute the steps of the method as described above by calling the computer program stored in the memory.

[0085] In addition, another embodiment of the present invention further provides a computer-readable storage medium for storing a computer program for robust optimization of distributed photovoltaic power output. When the computer program runs on a computer, it executes the steps of the method described above.

[0086] The forms of common computer-readable storage media generally include: floppy disks, flexible disks, hard disks, magnetic tapes, any other magnetic media, CD-ROMs, any other optical media, punch cards, paper tapes, any other physical media with a pattern of holes, random access memories (RAMs), programmable read-only memories (PROMs), erasable programmable read-only memories (EPROMs), flash erasable programmable read-only memories (FLASH-EPROMs), any other memory chips or cartridges, or any other media readable by a computer. The instructions can further be transmitted or received by a transmission medium. The term transmission medium can include any tangible or intangible medium that can be used to store, encode, or carry instructions for execution by a machine, and includes digital or analog communication signals or the intangible medium associated with facilitating the communication of the above instructions. The transmission medium includes coaxial cables, copper wires, and optical fibers, which include the wires of a bus used to transmit a computer data signal.

[0087] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program code. The solutions in the embodiments of the present application can be implemented in various computer languages. For example, object-oriented programming languages such as Java and interpreted scripting languages such as JavaScript can be used.

[0088] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to generate a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices generate for implementing in the process Figure 1 a process or multiple processes and / or blocks Figure 1means for the functions specified in one or more blocks.

[0089] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction means that implements the functions specified in one Figure 1 or more processes and / or blocks Figure 1 means for the functions specified in one or more blocks.

[0090] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus, such that a series of operational steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one Figure 1 or more processes and / or blocks Figure 1 means for the functions specified in one or more blocks.

[0091] Although the preferred embodiments of the present application have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present application.

[0092] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

[0093] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for robust optimization of distributed photovoltaic output, characterized in that: Includes the following: A two-stage dispatch strategy is adopted and combined with flexible resource dispatch and demand response on the demand side to construct the objective function of the overall operation cost of the microgrid and set the constraints of the objective function; The improved transit search algorithm is used to optimize the objective function and obtain the optimal solution. The QR algorithm is used to generate a confidence interval based on the optimal solution prediction, and the credibility of the optimal solution is judged based on whether the optimal solution falls within the confidence interval.

2. The method for robust optimization of distributed photovoltaic output according to claim 1, characterized in that: The objective function of the overall operating cost of the microgrid is: Among them, Y one represents the cost target of the pre-scheduling stage, Y two represents the cost target of the rescheduling phase, and denote the startup cost and shutdown cost of the microturbine MT, respectively, and They represent the startup cost and shutdown cost of distributed generation DG, d MT ,d DG ,d BS and d grid They represent the operating cost of the micro-turbine MT, the operating cost of the distributed generation DG, the charging and discharging cost of the energy storage system, and the power transaction cost between the microgrid and the large grid, respectively. U represents the uncertainty set, u represents the uncertainty variable, including photovoltaic power generation and load demand, x represents the decision variable set in the pre-dispatch stage, y represents the decision variable set in the re-dispatch stage, Q represents the total power reduction in response to the demand side, and q t represents the power reduction on the demand side per unit time, T represents the time length, and F(t) represents the price cost of calling the demand side flexibility resources. E1 and E2 represent the unmet demand of the demand side load for valley filling demand response and peak shaving demand response, respectively. f1 and f2 represent the unit compensation cost of valley filling demand response and peak shaving demand response, respectively. DR represents the cost of implementing demand side response. N DR represents the number of users responding on the demand side, v represents the critical power value that triggers the compensation price, S DR Indicates the compensation amount above the critical power value, L DR represents the compensation amount below the critical power value, z a and z b Respectively represent the compensation ratio coefficients above and below the critical power value.

3. The method for robust optimization of distributed photovoltaic output according to claim 1, characterized in that: The process of optimizing and solving the objective function using the improved transit search algorithm includes the following contents: In the development stage of the transit search algorithm, the sine-cosine mutation strategy is adopted to update the current solution. The update expression is: Among them, f E represents the current solution obtained by the transit search algorithm during the development phase, f′ E represents the updated solution, r1, r2, r3 and r4 are all random values, Represents the j-dimensional position of the optimal solution after t iterations.

4. The method for robust optimization of distributed photovoltaic output according to claim 3, characterized in that: The size of the random value r1 decreases as the number of iterations increases.

5. The method for robust optimization of distributed photovoltaic output according to claim 3, characterized in that: The process of optimizing and solving the objective function using the improved transit search algorithm includes the following contents: In the neighborhood stage of the transit search algorithm, the characteristics of the Cauchy distribution are used to update the current solution. The update expression is: Among them, f p represents the current solution obtained by the transit search algorithm in the neighborhood stage, θ represents a random value drawn from the Cauchy distribution, and f p ′ represents the updated solution.

6. The method for robust optimization of distributed photovoltaic output according to claim 1, characterized in that: The process of using the QR algorithm to generate a confidence interval based on the optimal solution prediction and judging the credibility of the optimal solution according to whether the optimal solution falls within the confidence interval includes the following: The QR algorithm is used to predict the optimal solution based on the estimated values ​​at the 0.05 quantile and the 0.95 quantile to generate a 90% confidence interval. If the optimal solution falls within the 90% confidence interval, the optimal solution is determined to be credible.

7. The method for robust optimization of distributed photovoltaic output according to claim 1, characterized in that: The constraints of the objective function include the power supply and demand balance constraints, which are expressed as follows: in, It represents the total actual power output of thermal power units. It represents the sum of the actual power generation output of wind turbines and photovoltaic units, S W Indicates the actual power generation output of the wind turbine, S P Indicates the actual power generation output of the photovoltaic unit. and They represent the total charging power and total charging power of the energy storage node respectively, Represents the total required power on the load side, S load , S Z , S f and S d They represent the benchmark load power, valley-filling demand response power, peak-shaving demand response power and power-deficit load power respectively.

8. A distributed photovoltaic power output robustness optimization system, characterized in that: include: The objective function construction module is used to adopt a two-stage dispatch strategy and combine the flexible resource dispatch and demand response on the demand side to construct the objective function of the overall operation cost of the microgrid and set the constraint conditions of the objective function; The optimization solution module is used to optimize and solve the objective function using an improved transit search algorithm to obtain the optimal solution; The credibility judgment module is used to generate a confidence interval based on the optimal solution prediction using the QR algorithm, and judge the credibility of the optimal solution according to whether the optimal solution falls within the confidence interval.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the method according to any one of claims 1 to 7 by calling the computer program stored in the memory.

10. A computer-readable storage medium for storing a computer program for robust optimization of distributed photovoltaic output, characterized in that: When the computer program is run on a computer, the steps of the method according to any one of claims 1 to 7 are executed.

Images