Optimized scheduling method, device and equipment for micro-grid equipment and storage medium

Abstract

The invention relates to an optimal scheduling method and device for micro-grid equipment, equipment and a storage medium. According to the main technical scheme, the method comprises the following steps: constructing a wind-light combined output model according to a wind-light output edge distribution model and a combined distribution model, and generating a typical day scene; according to the wind-solar combined output model, the upper-layer target function model of the equipment capacity and the upper-layer constraint conditions, an upper-layer planning model is constructed, and the capacity optimal solution of the equipment is obtained through whale optimization solution; constructing a lower-layer planning model according to the typical day scene, the lower-layer objective function model of the equipment power and the lower-layer constraint condition, and solving through dual planning to obtain an optimal power solution of the equipment; and according to the capacity optimal solution, the power optimal solution and the dual gap model, completing the optimal scheduling operation of the micro-grid equipment. According to the application, upper and lower layer collaborative optimization can be realized, and the effect of improving the operation stability and economical efficiency of the micro-grid system is achieved.

Description

Technical Field

[0001] This application relates to the field of microgrid technology, and in particular to an optimized scheduling method, apparatus, device, and storage medium for microgrid equipment. Background Technology

[0002] With the vigorous development of new energy technologies, a large number of distributed power sources and energy storage systems have begun to be connected to the distribution network, transforming it from a traditional unidirectional power flow system into a complex bidirectional system. While the extensive use of distributed energy helps improve the system's environmental benefits, the volatility of its output and its dependence on the randomness of the operating environment pose significant challenges to the safe and stable operation of the distribution network system. The development of microgrid technology allows distributed power sources, equipment, and controllable loads to form a small, self-managed and controlled system, thus largely mitigating the negative effects of a large number of distributed power sources on the distribution network. Therefore, the optimized scheduling of microgrids is particularly important.

[0003] In traditional implementations, most optimization scheduling methods use independent probability distributions (such as Weibull and Beta distributions) to describe wind and solar power output or only consider one of the two cases, without taking into account the uncertainty and correlation of wind and solar power output. In addition, mixed integer programming is used to solve the upper and lower level planning problems.

[0004] However, the above methods all have certain drawbacks: using independent probability distributions can easily overlook the nonlinear correlation between wind power and photovoltaic power under extreme weather conditions; considering only one of the two cases, wind power and photovoltaic power, cannot fully cover the intraday fluctuations of load in actual operation; and in solving the bi-level programming problem, if mixed integer programming is used, computational explosion can easily occur when dealing with high-dimensional discrete variables. Summary of the Invention

[0005] Based on this, this application provides an optimized scheduling method, device, equipment, and storage medium for microgrid equipment. By constructing a wind-solar combined output model, an upper-level planning model, and a lower-level planning model, and solving them through whale optimization and dual programming, a typical daily scenario model, the optimal capacity solution for the equipment, and the optimal power solution for the equipment are obtained. This enables the optimized scheduling operation of microgrid equipment, achieving coordinated optimization between the upper and lower levels, and improving the stability and economy of microgrid system operation.

[0006] Firstly, an optimized scheduling method for microgrid equipment is provided, the method comprising: Based on the edge distribution model and joint distribution model of wind and solar power output, a wind and solar joint power output model is constructed, and typical daily scenes are generated. Based on the combined wind and solar power output model, the upper-level objective function model of equipment capacity, and the upper-level constraints, an upper-level planning model is constructed, and the optimal solution for equipment capacity is obtained through whale optimization. Based on the typical daily scenario, the lower-level objective function model of equipment power, and the lower-level constraints, a lower-level planning model is constructed, and the optimal solution for equipment power is obtained by solving the dual programming problem. Based on the capacity-optimal solution, the power-optimal solution, and the dual gap model, the optimized scheduling operation of microgrid equipment is completed.

[0007] According to one feasible method in an embodiment of this application, a wind-solar combined output model is constructed based on the edge distribution model and the combined distribution model of wind and solar power output, and a typical daily scene is generated, including: Based on the combined wind and solar power output, the original power output scenario is generated; The original output scenario is reduced in dimensionality to obtain a dimensionality-reduced scenario. Clustering operations are performed on the dimensionality-reduced scenarios to generate typical daily scenarios.

[0008] According to one feasible method in the embodiments of this application, an upper-level planning model is constructed based on the wind-solar combined output model, the upper-level objective function model of equipment capacity, and the upper-level constraints. The optimal solution for equipment capacity is then obtained through whale optimization, including: Based on the combined wind and solar power output model, the upper-level objective function model, and the upper-level constraints, an upper-level planning model is constructed. Based on the upper-level planning model and fitness function model, the upper-level planning model is solved using whale optimization to obtain the capacity-optimal solution.

[0009] According to one achievable method in the embodiments of this application, the upper-level constraints include any one or more of the following: operating cost constraints of microgrid devices, output power constraints of microgrid devices, power balance constraints of microgrid devices, and lifetime loss constraints of microgrid devices.

[0010] According to one achievable method in an embodiment of this application, whale optimization includes a shrinking encirclement phase and a spiral update phase. Based on the upper-level planning model and the fitness function model, whale optimization is used to solve the upper-level planning model to obtain the capacity-optimal solution, including: Based on the upper-level programming model, fitness function model, and hybrid crossover model, the upper-level programming model is iteratively updated during the shrinking and encircling phase to obtain the capacity-optimal solution; or, Based on the upper-level planning model, fitness function model, and hybrid crossover model, the upper-level planning model is iteratively updated in a spiral update stage to obtain the capacity-optimal solution.

[0011] According to one achievable method in an embodiment of this application, dual programming includes dual transformation and optimization solution. Based on a typical daily scenario, a lower-level objective function model for device power, and lower-level constraints, a lower-level programming model is constructed. The optimal solution for device power is obtained through dual programming, including: Based on typical daily scenarios, lower-level objective function models, and lower-level constraints, a lower-level planning model is constructed. Based on the lower-level programming model, the lower-level constraints are transformed to obtain dual variables; Based on the lower-level programming model, the dual variables are solved to obtain the optimal power solution.

[0012] According to one achievable method in the embodiments of this application, the lower-level constraints include any one or more of the constraints on wind and solar power output and load use intervals, as well as the power balance constraints of microgrid equipment.

[0013] Secondly, an optimized scheduling device for microgrid equipment is provided, the device comprising: Typical day scene unit is used to construct a wind and solar power output model and generate typical day scenes based on the edge distribution model and joint distribution model of wind and solar power output; The capacity optimal solution unit is used to construct the upper-level planning model based on the wind and solar combined output model, the upper-level objective function model of equipment capacity, and the upper-level constraints, and to obtain the capacity optimal solution of the equipment through whale optimization. The power optimal solution unit is used to construct a lower-level planning model based on the lower-level objective function model and lower-level constraints of typical daily scenarios and equipment power, and to obtain the power optimal solution of the equipment through dual programming. The optimization scheduling unit is used to perform optimized scheduling operations on microgrid equipment based on the capacity optimal solution, the power optimal solution, and the dual gap model.

[0014] Thirdly, a computer device is provided, comprising: At least one processor; and A memory that is communicatively connected to at least one processor; wherein, The memory stores computer instructions that can be executed by at least one processor to enable the at least one processor to perform the methods involved in the first aspect above.

[0015] Fourthly, a computer-readable storage medium is provided, having stored thereon computer instructions, characterized in that the computer instructions are used to cause a computer to perform the methods involved in the first aspect above.

[0016] According to the technical content provided in the embodiments of this application, a wind-solar combined output model is constructed based on the edge distribution model and joint distribution model of wind and solar power output, and a typical daily scenario is generated. Based on the wind-solar combined output model, the upper-level objective function model of equipment capacity, and the upper-level constraints, an upper-level planning model is constructed, and the optimal capacity solution for the equipment is obtained through whale optimization. Based on the typical daily scenario, the lower-level objective function model of equipment power, and the lower-level constraints, a lower-level planning model is constructed, and the optimal power solution for the equipment is obtained through dual programming. Based on the optimal capacity solution, the optimal power solution, and the dual gap model, the optimized scheduling operation of the microgrid equipment is completed. The above operations, by constructing the wind-solar combined output model, the upper-level planning model, and the lower-level planning model, and solving them through whale optimization and dual programming, obtain the typical daily scenario model, the optimal capacity solution for the equipment, and the optimal power solution for the equipment, thereby completing the optimized scheduling operation of the microgrid equipment, achieving upper and lower level collaborative optimization, and improving the stability and economy of the microgrid system operation. Attached Figure Description

[0017] Figure 1 This is a system block diagram of an optimized scheduling method for microgrid devices in one embodiment; Figure 2 This is a flowchart illustrating an optimized scheduling method for microgrid devices in one embodiment; Figure 3 This is a schematic diagram of the original power output scenario in an optimized scheduling method for microgrid devices in one embodiment. Figure 4 This is a schematic diagram of a typical daily scenario in an optimized scheduling method for microgrid devices in one embodiment; Figure 5 This is a schematic diagram of the preferred process of an optimized scheduling method for microgrid devices in one embodiment; Figure 6 This is a structural block diagram of an optimized scheduling device for a microgrid in one embodiment; Figure 7 This is a schematic structural diagram of a computer device in one embodiment. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0019] For ease of understanding, the system to which this application applies is first described. This application provides a microgrid optimization scheduling method that can be applied to computer equipment, which may include a terminal or a server. Specifically, the computer equipment constructs a wind-solar combined output model and generates typical daily scenarios based on the edge distribution model and joint distribution model of wind and solar power output; it constructs an upper-level planning model based on the wind-solar combined output model, the upper-level objective function model of equipment capacity, and upper-level constraints, and obtains the optimal capacity solution for the equipment through whale optimization; it constructs a lower-level planning model based on the typical daily scenarios, the lower-level objective function model of equipment power, and lower-level constraints, and obtains the optimal power solution for the equipment through dual programming; based on the optimal capacity solution, the optimal power solution, and the dual gap model, it completes the optimization scheduling operation of the microgrid equipment. The terminal can be, but is not limited to, a microgrid, and the server can be a standalone server or a server cluster composed of multiple servers.

[0020] In one embodiment, such as Figure 1 , Figure 2 As shown, an optimized scheduling method for microgrid devices is provided. Taking the application of this method to computer equipment as an example, the method includes the following steps: Step S101: Based on the edge distribution model and joint distribution model of wind and solar power output, construct a wind and solar joint power output model and generate a typical daily scene.

[0021] Specifically, to characterize the probability distribution of active power output from wind and solar power sources, a kernel density estimation method can be used to construct their probability distribution model, i.e., the marginal distribution model. The marginal distribution models for both wind and solar power sources can be derived using... express, The specific expression can be represented as follows: ; in, This represents the target value of active power output of wind / solar power sources after standardization, i.e., the estimated value of active power output, covering the output range of wind / solar power sources during operation; Let represent the measured, normalized active power of the wind / photovoltaic power source at time i, and , n represents the total measured active power during the operating cycle of the wind / photovoltaic power source. The operating cycle can generally be set to 24 hours, with sampling every 15 minutes. For kernel function, Indicates bandwidth.

[0022] Here, wind power and photovoltaic power are referred to as wind-solar for short. For estimating the marginal distribution of wind and solar power output, a Gaussian kernel function can be chosen; and due to bandwidth... The selection of ... Compared with the true marginal distribution model The deviation is used to determine this. The specific kernel function expression and bandwidth are also considered. The expression can be represented as follows: ; Here, E represents the weight matrix, which can be preset for comprehensive statistical analysis of landscape characteristics. It should be noted that, due to the true edge distribution model... It is derived from measured data, so there is no specific expression. Generally, it resembles a beta distribution when representing photovoltaic power and a Weibull distribution when representing wind power.

[0023] Since the output of wind and solar power is correlated to a certain extent, after obtaining the marginal distribution model using the kernel function, a joint distribution model can be used to describe the correlation and uncertainty of wind and solar power output. The joint distribution model can be represented by C, and its specific expression is as follows: ; Where u1 and u2 represent the output probability values ​​of the wind power edge distribution model and the photovoltaic edge distribution model, respectively; This represents a correlation parameter, which can be preset. The larger the value, the stronger the positive correlation between the variables.

[0024] Based on the aforementioned edge distribution model and joint distribution model, the wind-solar joint output model is constructed, and then typical daily scenes can be generated based on the wind-solar joint output model.

[0025] Step S103: Based on the wind-solar combined output model, the upper-level objective function model of equipment capacity, and the upper-level constraints, construct the upper-level planning model, and obtain the optimal solution for equipment capacity through whale optimization.

[0026] Here, the main objective of the upper-level planning model is to optimize the device capacity in the microgrid, thereby minimizing the total lifecycle cost of device operation. Specifically, the upper-level objective function model for device capacity can be represented by F, and its specific expression is as follows: ; Where, r p This represents the discount rate, calculated uniformly for different equipment; r i Indicates the lifespan of different devices; c i This represents the unit capacity cost of different equipment; all of the aforementioned variables can be preset in advance. E i This indicates different device capacities, which need to be optimized. This represents the operating cost, derived from the lower-level planning model. Different devices can be represented, including but not limited to wind turbines, which can be represented by WT; photovoltaics can be represented by PV; generators can be represented by G; and energy storage can be represented by bat.

[0027] Based on the combined wind and solar power output model, the aforementioned upper-level objective function model, and the upper-level constraints, the upper-level planning model is constructed, and the optimal solution for equipment capacity is obtained through whale optimization.

[0028] Step S105: Based on the typical daily scenario, the lower-level objective function model of equipment power, and the lower-level constraints, construct the lower-level planning model, and obtain the optimal solution for equipment power through dual programming.

[0029] Here, the construction of the lower-level planning model aims to improve power supply reliability and renewable energy absorption rate. Specifically, the lower-level objective function model can be... The specific expression can be represented as follows: ; in, The penalty coefficient for load shedding can be preset in advance; The penalty coefficient for abandoning wind and solar power can be preset in advance; This represents the load shedding power of a typical day d during time period t; This represents the wind curtailment power on a typical day d during time period t; This represents the abandoned power of light on a typical day d during time period t; , as well as All of these need to be optimized; C represents the number of typical daily scenarios.

[0030] Based on typical daily scenarios, the aforementioned lower-level objective function model, and lower-level constraints, the lower-level planning model is constructed, and the optimal power solution for the equipment is obtained through dual programming.

[0031] Step S107: Based on the capacity optimal solution, power optimal solution, and dual gap model, complete the optimized scheduling operation of microgrid equipment.

[0032] Specifically, the consistency between the upper and lower planning can be determined using the dual gap model, and the specific expression can be represented as follows: ; in, Indicates the duality gap; This represents the capacity-optimal solution; This represents the optimal power solution; This indicates the goals of the higher-level planning, namely... corresponding The target value obtained by solving; Indicates the lower-level planning objectives, namely The corresponding target value.

[0033] Here, a convergence threshold can be set. =0.5%, when If the upper and lower level planning is considered to have converged, the iteration is terminated; otherwise, the power optimal solution is obtained. Feedback is sent to the upper-level planning model for iterative updates until the optimized scheduling of microgrid equipment is completed.

[0034] As can be seen, the embodiments of this application construct a combined wind and solar power output model based on the edge distribution model and the joint distribution model of wind and solar power output, and generate typical daily scenarios; construct an upper-level planning model based on the combined wind and solar power output model, the upper-level objective function model of equipment capacity, and the upper-level constraints, and obtain the optimal capacity solution of the equipment through whale optimization; construct a lower-level planning model based on the typical daily scenarios, the lower-level objective function model of equipment power, and the lower-level constraints, and obtain the optimal power solution of the equipment through dual programming; and complete the optimized scheduling operation of microgrid equipment based on the optimal capacity solution, the optimal power solution, and the dual gap model. The aforementioned operations, by constructing the combined wind and solar power output model, the upper-level planning model, and the lower-level planning model, and solving them through whale optimization and dual programming, obtain the typical daily scenario model, the optimal capacity solution of the equipment, and the optimal power solution of the equipment, thereby completing the optimized scheduling operation of microgrid equipment, achieving upper and lower level collaborative optimization, and improving the stability and economy of microgrid system operation.

[0035] The following is a detailed description of each step in the above method flow. First, step 101, namely "constructing a wind-solar combined output model and generating a typical daytime scene based on the edge distribution model and combined distribution model of wind and solar power output", will be described in detail with reference to the embodiment.

[0036] Based on the wind-solar combined power output model, the original power output scenario is generated; the original power output scenario is then subjected to dimensionality reduction to obtain a dimensionality-reduced scenario; the dimensionality-reduced scenario is then subjected to clustering to generate a typical daily scenario.

[0037] Specifically, refer to Figure 3 Based on the combined wind and solar power model, the original power output scenario is generated. The value range will be... The marginal distribution model is divided into N equal intervals, and the expression for the i-th interval can be expressed as: , In interval I i Multiple values ​​are randomly selected from within the range, which can be represented by u, where u follows a uniform distribution. .

[0038] The inverse function can be obtained by solving the joint distribution model C. Let's assume we use... The specific expression is as follows: ; From the above expression, we can obtain The value of is the probability value of the edge distribution model for wind power and photovoltaics. After obtaining... Then, the inverse function of the marginal distribution model can be used, assuming that... This means that a solution can then be obtained. This yields the corresponding active power output estimates for wind power and photovoltaic power. The more u values ​​extracted, the more active power output estimates for wind power and photovoltaic power are obtained. Based on multiple active power output estimates for wind power and photovoltaic power, a large number of corresponding original output scenarios can be generated.

[0039] Reference Figure 4 The original output scenario model is then subjected to dimensionality reduction to obtain a dimensionality-reduced output scenario. Specifically, PCA can be used for dimensionality reduction, and the specific steps are as follows: ; Where m represents the total number of original output scenarios; n represents the time feature dimension of a single scenario, which is generally set to 96, that is, within 24 hours, it is sampled once every 15 minutes to obtain 96 time features; m*m scene covariance matrix; Let m*n be the scene matrix after centering, where X represents the m*n original scene matrix; express Feature mean vector; W k Let m*k represent the m*k matrix, which is composed of the feature vectors corresponding to the first k original power output scenarios; Y represents the k*n low-dimensional feature matrix after dimensionality reduction, thereby compressing the m*n matrix to a k*n matrix to obtain the dimensionality-reduced power output scenario.

[0040] Clustering is performed on the dimensionality-reduced output scenarios to generate typical daily scenarios. Specifically, for the dimensionality-reduced feature matrix Y, fuzzy C-means classification is used, and the specific expression can be represented as follows: ; Where J represents the clustering objective function; c represents the preset number of clusters; Y j (j=1,2,…,k) represents the j-th dimension-reduced scenario in PCA dimensionality reduction, V i (i=1,2,…,c) represents the i-th cluster center; q represents the fuzzy coefficient; l represents the cluster center index; u ijThe membership degree is represented by the number of cluster centers. By adjusting the membership degree, the objective function J can be solved until convergence, and the c*n matrix of clustering results can be output, thus obtaining the typical daily scene.

[0041] The above-mentioned dimensionality reduction and clustering operations on the original output scenario model yield typical daily scenarios, thereby reducing the number of data features and solving problems such as high computational complexity and high risk of overfitting caused by high-dimensional data. At the same time, it improves visualization and model performance, as well as the effectiveness of data analysis and detection of abnormal scenarios.

[0042] The following describes in detail step S103, namely, "constructing an upper-level planning model based on the wind-solar combined output model, the upper-level objective function model of equipment capacity, and the upper-level constraints, and obtaining the optimal solution for equipment capacity through whale optimization," with reference to the embodiments.

[0043] Based on the combined wind and solar power output model, the upper-level objective function model, and the upper-level constraints, an upper-level planning model is constructed. Based on the upper-level planning model and the fitness function model, the upper-level planning model is solved through whale optimization to obtain the capacity-optimal solution.

[0044] Here, an upper-level planning model is constructed based on the wind-solar combined output model, the upper-level objective function model, and the upper-level constraints. Specifically, the upper-level constraints include, but are not limited to, the operating cost constraints of the microgrid equipment, the output power constraints of the microgrid equipment, the power balance constraints of the microgrid equipment, and the lifetime loss constraints of the microgrid equipment.

[0045] In one feasible approach, the operating cost constraint of microgrid devices The expression can be represented as follows: ; Where, π t,d Indicates time-of-use electricity pricing. , indicating hours; , representing the off-peak, flat, and peak periods for electricity prices, respectively; c fuel This indicates fuel cost, which can be preset; c i,om This represents the operation and maintenance cost coefficient, which can be preset in advance; P buy / sell This represents the power purchased / sold, which can be optimized; P g This represents the generator's output power, which can be optimized to obtain; P i This represents the estimated output power of the corresponding equipment. The estimated output power of wind power and photovoltaic power can be obtained through a wind-solar combined output model. The estimated output power of generators and energy storage is initially determined by configuring an upper and lower limit range of values, and can then be adjusted based on constraints.

[0046] The output power constraint of microgrid equipment can be expressed as follows: ; Among them, P ch / P dis Indicates the charging / discharging power of the energy storage device; η ch / η dis Indicates the energy storage charge / discharge efficiency; u ch / dis Represents the charging / discharging binary state variable (taking only 0 or 1, and...) ); Indicates the upper limit of energy storage capacity; P g Indicates generator power; Indicates the upper limit of generator power; Indicates the upper limit of power exchange in the distribution network; P buy / P sell Indicates the power purchased / sold; E bat E represents the amount of energy stored; bat,t This represents the amount of energy stored in hour t.

[0047] The power balance constraint of microgrid equipment can be expressed as follows: ; in, This indicates the estimated output power of wind power. This indicates the estimated output power of photovoltaic power. P represents the estimated output power of the generator; buy / P sell Indicates the power purchased / sold; P load Indicates load demand; ΔE bat This indicates the net charge and discharge amount of energy storage.

[0048] Here, in addition to satisfying the mutual exclusion of charging and discharging states and power constraints, the energy storage device also needs to quantify its lifespan loss. That is, the energy storage device has different loss rates in different cycle number ranges, and the depth of discharge of the device in different lifespans needs to be limited.

[0049] Specifically, the lifespan loss constraint of microgrid equipment is related to time... The constraint expression can be represented as follows: ; in, This represents the lifespan cost of the energy storage device at time t; k j and b j Let represent the slope and intercept of the j-th piecewise function; The cumulative number of charge-discharge cycles at time t represents the energy storage device; gj(t) represents a binary variable, where gj(t)=1 indicates that the j-th piecewise function is activated, and gj(t)=0 indicates that the term has no effect in the constraint; M is a maximal positive real number used to transform the piecewise function into a linear constraint; m represents the number of segments in the life cycle and discharge depth of the energy storage device; the total discharge depth D(t) is equal to the sum of the interval components dj(t), and lj and uj represent the upper and lower limits of the interval. By setting different lj and uj, shallow discharge is performed under low loss and deep discharge is performed under high loss to balance the operating requirements.

[0050] As shown above, the estimated power output of wind power and the estimated power output of photovoltaic power can be obtained through the wind-solar combined power output model. Then, based on the estimated power output of wind power, the estimated power output of photovoltaic power, the upper-level objective function model, and the constraints of the above-mentioned upper-level constraints, the upper-level planning model can be constructed.

[0051] Based on the upper-level planning model and fitness function model, the upper-level planning model is solved using whale optimization to obtain the capacity-optimal solution.

[0052] In one feasible approach, the upper-level planning model is iteratively updated in a shrinking encirclement phase based on the upper-level planning model, fitness function model, and hybrid crossover model to obtain the capacity-optimal solution; or, the upper-level planning model is iteratively updated in a spiral update phase based on the upper-level planning model, fitness function model, and hybrid crossover model to obtain the capacity-optimal solution.

[0053] Whale optimization includes a shrinking encirclement phase and a spiral update phase.

[0054] Here, when solving the upper-level planning model, we can first perform population initialization, where the combination of each individual x(t) is: Right now This represents the capacity combination of different devices. The objective function F in the upper-level planning model serves as the fitness model for the optimization algorithm, with a penalty function added to handle constraints. Specifically, the fitness function model can be... The specific expression can be represented as follows: ; Where F represents the upper-level objective function model; G represents the coefficient of the penalty function; i (x) represents the constraint violation; i represents different devices; n represents the total number of devices.

[0055] Preferably, in optimizing and updating the objective function, the upper-level programming model can be iteratively updated in a contraction and encirclement phase based on the upper-level programming model, fitness function model, and hybrid crossover model to obtain the capacity-optimal solution. The contraction and encirclement phase refers to the phase where the absolute value of the convergence factor is less than 1. Assuming the convergence factor is represented by A, i.e., |A| < 1. The specific expression used in the contraction and encirclement phase can be expressed as follows: ; in, a decreases linearly from 2 to 0; C represents the random coefficient, and r∈[0,1]; t represents the number of iterations; This represents the different combinations of device capacities at the t-th iteration; This indicates the current optimal solution, which is the optimal solution for the equipment's capacity.

[0056] Here, during the iteration process, a hybrid crossover model can be used for further optimization. The specific expression of the hybrid crossover model can be represented as follows: ; in, The x represents the latest individual in iteration t+1; β represents a normally distributed random number; x j and x k This represents elite individuals in the population. The selection rule is to select the top 20% of individuals based on their fitness function, and then randomly select two of these individuals for crossover.

[0057] Specifically, assuming that in the first iteration, the solution obtained from the shrinking encirclement phase can be used to obtain the solution individuals for the first iteration, and then the elite individuals obtained from the aforementioned solution individuals can be mixed and cross-pollinated to obtain better individuals, i.e. In the second iteration, it can be This process is repeated in the shrinking encirclement solution stage until the upper limit of the number of iterations is reached, at which point the algorithm converges, thus obtaining the optimal solution for the device's capacity.

[0058] Preferably, based on the upper-level planning model, fitness function model, and hybrid crossover model, the upper-level planning model is iteratively updated in a spiral update phase to obtain the optimal capacity solution for the equipment. The spiral update phase refers to the phase where the absolute value of the convergence factor is greater than or equal to 1, i.e. The specific expression used in the spiral update phase can be represented as follows: ; in, This represents the different combinations of device capacity at the (t+1)th iteration, i.e., the next iteration, which is generally achieved through an optimization approach. Indicates the helix radius. ; This represents the parameters of the spiral shape, generally speaking. ; Represents a random angle parameter. . This indicates the current optimal solution, which is the optimal solution for the equipment's capacity.

[0059] Here, during the iteration process, a hybrid cross model can also be used for further optimization. The specific process is the same as the shrinking and encircling stage, only the corresponding stage is different and the individual solutions obtained are different. However, the principle of the optimization process of the hybrid cross model is the same, so it will not be repeated here. Finally, after reaching the upper limit of the number of iterations, the algorithm converges and obtains the optimal solution for the device capacity.

[0060] The above operations construct an upper-level planning model with the goal of minimizing the total life cycle cost, introduce piecewise functions to describe the energy storage lifespan loss, and solve the equipment capacity combination through the adaptive whale algorithm, which can quickly find the optimal solution, thereby realizing the coordinated optimization of microgrid equipment capacity configuration and operation strategy.

[0061] Finally, the above step S105, namely "constructing a lower-level planning model and obtaining the optimal power solution of the equipment based on the lower-level objective function model and lower-level constraints of a typical daily scenario and equipment power", will be described in detail with reference to the embodiments.

[0062] Based on typical daily scenarios, lower-level objective function models, and lower-level constraints, a lower-level planning model is constructed. Based on the lower-level planning model, the lower-level constraints are transformed to obtain dual variables. Based on the lower-level planning model, the dual variables are solved to obtain the optimal power solution.

[0063] Dual programming includes dual transformation and optimization solution.

[0064] In one feasible approach, the lower-level constraints include, but are not limited to, constraints on wind and solar power output and load use areas, as well as power constraints on microgrid equipment.

[0065] Here, the power balance constraint of microgrid equipment can be expressed as follows: ; in, This indicates the estimated output power of wind power. This indicates the estimated output power of photovoltaic power. P represents the generator's output power; buy / P sell Indicates the power purchased / sold; P load Indicates load demand; ΔE bat This indicates the net charge and discharge amount of energy storage.

[0066] Here, the joint distribution function can be used to describe wind power P. WT,t,d and photovoltaic P PV,t,d The correlation of output power can be expressed as: Then, the expressions for wind and solar power output and load constrained by the interval can be expressed as follows: ; in, and These represent the predicted wind power and photovoltaic power output for the t-th time period on the c-th typical day, respectively. and ω represents the fluctuation of the predicted wind power and photovoltaic power output during the t-th time period on the c-th typical day; WT and ω PV These represent the uncertainty ratios for wind power and solar power, respectively. This represents the actual load during the t-th time period of the c-th typical day; ω represents the fluctuation of the actual load during the t-th time period on the c-th typical day; load This represents the load uncertainty ratio.

[0067] As can be seen from the above, based on the number of typical daily scenarios, the lower-level objective function model, and the aforementioned lower-level constraints, the construction of the lower-level planning model can be completed.

[0068] Based on the lower-level programming model, the lower-level constraints are transformed to obtain the dual variables. The specific expression can be represented as follows: ; in, Describes the objective function of the lower-level programming model. gradient; Represents the generator constraint matrix; Represents the energy storage constraint matrix; Represents the power constraint matrix of the microgrid; Indicates the generator power constraint threshold; Indicates the energy storage power constraint threshold; Indicates the power constraint threshold of the distribution network; Represents the power balance coefficient matrix; This represents a vector containing uncertainties related to wind, solar, and load. This represents the dual variable.

[0069] Based on the lower-level programming model, the dual variables are solved to obtain the optimal power solution. Here, the specific expression for solving the dual variables can be represented as follows: ; In the formula, ρ represents the objective function of the lower-level programming model; ρ represents the uncertainty penalty factor. and These represent the fluctuations in the predicted power output of wind power and photovoltaic power during the t-th time period on the c-th typical day, respectively. Q represents the fluctuation of the actual load in the t-th time period of the c-th typical day; Q4 represents the matrix semidefinite programming constraint matrix; H represents the uncertainty coupling matrix; This represents a vector containing uncertainties related to wind, solar, and load. Let represent the identity matrix. Using CPLEX, a semidefinite programming problem is solved, leading to the optimal power solution for the equipment. .

[0070] The above operations, when optimizing the lower-level planning model, take into account the reliability of power supply, ensure the minimum wind and solar power curtailment load, and thus ensure the accuracy of the optimal power solution obtained by the lower-level model.

[0071] Based on the implementation methods in the above embodiments, the following will be combined with... Figure 5 A preferred method flow provided in an embodiment of this application will be described by way of example. For instance... Figure 5 As shown, the method may include the following steps: Step S201: Construct a wind-solar combined output model based on the edge distribution model and the combined distribution model of wind and solar power output.

[0072] Step S202: Generate the original power output scenario based on the wind-solar combined power output model.

[0073] Step S203: Perform a dimensionality reduction operation on the original output scenario to obtain a dimensionality-reduced scenario.

[0074] Step S204: Perform clustering operations on the dimensionality-reduced scenario to generate typical daily scenarios.

[0075] Step S205: Construct the upper-level planning model based on the wind-solar combined output model, the upper-level objective function model, and the upper-level constraints.

[0076] Step S206: Based on the upper-level planning model, fitness function model, and hybrid crossover model, perform iterative updates of the upper-level planning model in the shrinking and encircling phase to obtain the capacity-optimal solution; or, based on the upper-level planning model, fitness function model, and hybrid crossover model, perform iterative updates of the upper-level planning model in the spiral update phase to obtain the capacity-optimal solution.

[0077] Step S207: Construct a lower-level planning model based on typical daily scenarios, the lower-level objective function model, and the lower-level constraints.

[0078] Step S208: Based on the lower-level programming model, transform the lower-level constraints to obtain dual variables.

[0079] Step S209: Based on the lower-level planning model, solve the dual variables to obtain the optimal power solution.

[0080] Step S210: Based on the capacity optimal solution, power optimal solution, and dual gap model, complete the optimized scheduling operation of microgrid equipment.

[0081] It should be understood that, although Figure 2 , Figure 5 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated in this application, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Furthermore, Figure 2 , Figure 5 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0082] Figure 6 This is a schematic diagram of the structure of an optimized scheduling device for microgrid equipment provided in an embodiment of this application. This device can be installed in a computer device to perform tasks such as... Figure 2 , Figure 5 The method flow is shown below. Figure 6 As shown, the device may include: a typical daytime scenario unit 301, a capacity optimal solution unit 303, a power optimal solution unit 305, and an optimization scheduling unit 307. The main functions of each component module are as follows: Typical day scene unit 301 is used to construct a wind and solar power output model and generate typical day scenes based on the edge distribution model and joint distribution model of wind and solar power output. The capacity optimal solution unit 303 is used to construct an upper-level planning model based on the wind and solar combined output model, the upper-level objective function model of equipment capacity, and the upper-level constraints, and to obtain the capacity optimal solution of the equipment through whale optimization. The power optimal solution unit 305 is used to construct a lower-level planning model based on the lower-level objective function model and lower-level constraints of typical daily scenarios and equipment power, and to obtain the power optimal solution of the equipment through dual programming. The optimization scheduling unit 307 is used to complete the optimization scheduling operation of microgrid equipment based on the capacity optimal solution, the power optimal solution and the dual gap model.

[0083] In one embodiment, the typical daytime scene unit 301 is further used for: Based on the combined wind and solar power model, the original power output scenario is generated. The original output scenario is reduced in dimensionality to obtain a dimensionality-reduced scenario. Clustering operations are performed on the dimensionality-reduced scenarios to generate typical daily scenarios.

[0084] In one embodiment, the capacity-optimized solution unit 303 is further configured to: Based on the combined wind and solar power output model, the upper-level objective function model, and the upper-level constraints, an upper-level planning model is constructed. Based on the upper-level planning model and fitness function model, the upper-level planning model is solved using whale optimization to obtain the capacity-optimal solution.

[0085] In one embodiment, the capacity-optimized solution unit 303 is further configured to: The upper-level constraints include any one or more of the following: operating cost constraints of microgrid equipment, output power constraints of microgrid equipment, power balance constraints of microgrid equipment, and lifetime loss constraints of microgrid equipment.

[0086] In one embodiment, whale optimization includes a shrinking encirclement phase and a spiral update phase, and the capacity-optimal solution unit 303 is further used for: Based on the upper-level programming model, fitness function model, and hybrid crossover model, the upper-level programming model is iteratively updated during the shrinking and encircling phase to obtain the capacity-optimal solution; or, Based on the upper-level planning model, fitness function model, and hybrid crossover model, the upper-level planning model is iteratively updated in a spiral update stage to obtain the capacity-optimal solution.

[0087] In one embodiment, dual programming includes dual transformation and optimization solution, and the power optimal solution unit 305 is further used for: Based on typical daily scenarios, lower-level objective function models, and lower-level constraints, a lower-level planning model is constructed. Based on the lower-level programming model, the lower-level constraints are transformed to obtain dual variables; Based on the lower-level programming model, the dual variables are solved to obtain the optimal power solution.

[0088] In one embodiment, the power optimal solution unit 305 is further configured to: The lower-level constraints include any one or more of the constraints on wind and solar power output and load use areas, as well as the power balance constraints of microgrid equipment.

[0089] The same or similar parts among the above embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments.

[0090] It should be noted that the embodiments of this application may involve the use of user data. In practical applications, user-specific personal data may be used in the scheme described herein within the scope permitted by applicable laws and regulations, provided that it complies with the applicable laws and regulations of the country (e.g., explicit consent from the user, actual notification to the user, explicit authorization from the user, etc.).

[0091] According to embodiments of this application, this application also provides a computer device and a computer-readable storage medium.

[0092] like Figure 7 The diagram shown is a block diagram of a computer device according to an embodiment of this application. The term "computer device" is intended to represent various forms of digital computers or mobile devices. The digital computer may include a desktop computer, a portable computer, a workbench, a personal digital assistant, a server, a mainframe computer, and other suitable computers. The mobile device may include a tablet computer, a smartphone, a wearable device, etc.

[0093] like Figure 7 As shown, device 400 includes a computing unit 401, a ROM 402, a RAM 403, a bus 404, and an input / output (I / O) interface 405. The computing unit 401, ROM 402, and RAM 403 are interconnected via the bus 404. The input / output (I / O) interface 405 is also connected to the bus 404.

[0094] The computing unit 401 can execute various processes in the method embodiments of this application according to computer instructions stored in the read-only memory (ROM) 402 or computer instructions loaded from the storage unit 408 into the random access memory (RAM) 403. The computing unit 401 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. The computing unit 401 can include, but is not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. In some embodiments, the methods provided in the embodiments of this application can be implemented as computer software programs, which are tangibly contained in a computer-readable storage medium, such as the storage unit 408.

[0095] RAM 403 can also store various programs and data required for the operation of device 400. Part or all of the computer program can be loaded and / or installed on device 400 via ROM 402 and / or communication unit 409.

[0096] The input unit 406, output unit 407, storage unit 408, and communication unit 409 in device 400 can be connected to I / O interface 405. The input unit 406 can be, for example, a keyboard, mouse, touchscreen, or microphone; the output unit 407 can be, for example, a display, speaker, or indicator light. Device 400 can exchange information and data with other devices through the communication unit 409.

[0097] It should be noted that the device may also include other components necessary for normal operation. It may also include only the components necessary for implementing the solution of this application, without necessarily including all the components shown in the figures.

[0098] Various implementations of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SOCs), payload programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof.

[0099] The computer instructions used to implement the methods of this application may be written in any combination of one or more programming languages. These computer instructions may be provided to the computing unit 401 such that when executed by the computing unit 401, such as a processor, the computer instructions cause the execution of the steps involved in the embodiments of the methods of this application.

[0100] The computer-readable storage medium provided in this application can be a tangible medium that can contain or store computer instructions for performing the steps involved in the method embodiments of this application. The computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, and other forms of storage media.

[0101] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. An optimized scheduling method for microgrid equipment, characterized in that, The method comprises: According to the edge distribution model and the joint distribution model of wind and light output, a wind and light joint output model is constructed, and a typical day scene is generated; According to the wind and light joint output model, the upper layer objective function model of device capacity and the upper layer constraint condition, an upper layer planning model is constructed, and the capacity optimal solution of the device is obtained by whale optimization solution; According to the typical day scene, the lower layer objective function model of device power and the lower layer constraint condition, a lower layer planning model is constructed, and the power optimal solution of the device is obtained by dual programming solution; According to the capacity optimal solution, the power optimal solution and the dual gap model, the optimization scheduling operation of the microgrid device is completed.

2. The method of claim 1, wherein, According to the edge distribution model and the joint distribution model of wind and light output, a wind and light joint output model is constructed, and a typical day scene is generated, comprising: According to the wind and light joint output model, an original output scene is generated; The dimensionality reduction operation is performed on the original output scene to obtain a reduced dimension scene; The reduced dimension scene is clustered to generate the typical day scene.

3. The method of claim 1, wherein, According to the wind and light joint output model, the upper layer objective function model and the upper layer constraint condition, the upper layer planning model is constructed, and the capacity optimal solution of the device is obtained by whale optimization solution, comprising: According to the wind and light joint output model, the upper layer objective function model and the upper layer constraint condition, the upper layer planning model is constructed; According to the upper layer planning model and the fitness function model, the upper layer planning model is solved by whale optimization to obtain the capacity optimal solution.

4. The method of claim 3, wherein, The upper layer constraint condition comprises any one or more of the operation cost constraint of the microgrid device, the output power constraint of the microgrid device, the power balance constraint of the microgrid device and the life loss constraint of the microgrid device.

5. The method of claim 4, wherein, The whale optimization comprises a shrinkage surrounding stage and a spiral update stage, and the capacity optimal solution is obtained by solving the upper layer planning model by whale optimization according to the upper layer planning model and the fitness function model, comprising: According to the upper layer planning model, the fitness function model and the hybrid crossover model, the upper layer planning model is iteratively updated in the shrinkage surrounding stage to obtain the capacity optimal solution; or According to the upper layer planning model, the fitness function model and the hybrid crossover model, the upper layer planning model is iteratively updated in the spiral update stage to obtain the capacity optimal solution.

6. The method of claim 2, wherein, The dual programming comprises dual transformation and optimization solution, and the power optimal solution of the device is obtained by dual programming solution according to the typical day scene, the lower layer objective function model of device power and the lower layer constraint condition, comprising: According to the typical day scene, the lower layer objective function model and the lower layer constraint condition, the lower layer planning model is constructed; According to the lower layer planning model, the dual variable is obtained by transforming the lower layer constraint condition; According to the lower layer planning model, the power optimal solution is obtained by programming solution of the dual variable.

7. The method of claim 6, wherein, The lower layer constraint conditions include any one or more of wind and light output and load interval constraints and power balance constraints of micro-grid equipment.

8. An optimal scheduling device of a microgrid device, characterized in that, The device comprises: a typical day scenario unit configured to construct a wind and light joint output model according to an edge distribution model of wind and light output and a joint distribution model, and to generate a typical day scenario; a capacity optimal solution unit configured to construct an upper layer planning model according to the wind and light joint output model, an upper layer objective function model of equipment capacity and upper layer constraint conditions, and to obtain a capacity optimal solution of equipment by whale optimization solution; a power optimal solution unit configured to construct a lower layer planning model according to the typical day scenario, a lower layer objective function model of equipment power and lower layer constraint conditions, and to obtain a power optimal solution of equipment by dual programming solution; an optimal scheduling unit configured to complete optimal scheduling operation of micro-grid equipment according to the capacity optimal solution, the power optimal solution and a dual gap model.

9. A computer device, comprising: at least one processor; and a memory connected in communication with the at least one processor; wherein the memory stores computer instructions executable by the at least one processor, and the computer instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.

10. A computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are used to enable the computer to perform the method of any one of claims 1-7.

Images