Electric power system multi-resource collaborative planning method and system based on attribution analysis

Abstract

The invention relates to a power system multi-resource collaborative planning method based on attribution analysis, and belongs to the technical field of power system planning. Comprising the steps of constructing a target function by taking the minimum supply and demand mismatching degree as a target based on the load demand of a power system and the output of each unit, and constructing a planning model based on set constraints and the target function; solving the planning model based on the multiple groups of boundary condition values of the planning model to obtain corresponding planning schemes; analyzing to obtain quantitative expressions of the influence degrees of the boundary conditions and the constraints on the planning scheme; and corresponding boundary condition values are adjusted based on quantitative representation, the planning models are solved, and the planning scheme with the lowest planning cost in solving results is selected as the optimal planning scheme. According to the method, the key factors influencing the planning scheme are determined by performing attribution analysis on the electric power planning, and the optimal planning scheme is determined by adjusting the key factors, so that the scientificity and the economical efficiency of the planning scheme are further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system planning, and in particular, to a multi-resource collaborative planning method and system for power systems based on attribution analysis. Background Art

[0002] In the process of the green and low-carbon transformation of energy, the high-proportion access of new energy sources such as wind power and photovoltaic power has become an inevitable development trend in the new power system form. In the transformation process of the new power system, the coordinated development of multi-time-scale energy storage is an effective way to ensure the balance between supply and demand and the consumption of new energy. Therefore, it is necessary to make resource planning decisions based on the coordinated operation of multiple types of technologies, which poses more stringent requirements for power system planning methods.

[0003] Traditional planning models are all "black box", unable to explain the physical logic of the internal decision-making of the model. Although the planning decision-makers can obtain the optimal solution in the sense of the objective function, that is, under the premise of clear boundary conditions and constraint conditions, determine the optimal planning scheme of the power system.

[0004] However, since traditional planning models cannot clearly extract the dominant factors driving the optimization results from the model and cannot explore the decision-making logic inside the planning model, when facing the differences in the demand for different types of resources in different regions (which may be regions with a small geographical scope such as a park or a city, or regions with a large geographical scope such as the whole province or across provinces) or different-scale power grids (such as microgrids or large-scale power grids), there is a lack of a basis for formulating scientific and reasonable adjustments to boundary conditions and constraint conditions for each region; on the other hand, when facing the different demands for various resources at different planning and construction stages, there is also a lack of a basis for boundary conditions and constraint conditions for the different development characteristics at each stage (for example, the demand for different types of energy storage resources is different at different planning stages, and the strictness of carbon emission requirements is different at different planning stages). As a result, traditional planning models cannot effectively provide scientific support for power system planning decisions in different regions, different-scale power grids, and different regions at different planning and construction stages. Summary of the Invention

[0005] In view of the above analysis, the present invention aims to disclose a multi-resource collaborative planning method and system for power systems based on attribution analysis. When making power system planning decisions, through attribution analysis, determine the influence degree of each boundary condition or constraint condition on the planning scheme respectively, adjust the boundary conditions corresponding to the boundary conditions or constraint conditions that have a great influence on the planning scheme according to the different influence degrees, and determine the optimal power system planning decision based on the adjusted boundary conditions.

[0006] On the one hand, the present invention provides a multi-resource collaborative planning method for power systems based on attribution analysis, which specifically includes the following steps:

[0007] Based on the load demand of the power system and the output of each unit, a target function is constructed with the lowest supply-demand mismatch degree as the goal. Based on the power balance constraint, line power flow constraint, output constraint of each unit, low-carbon constraint and the target function, a planning model is constructed;

[0008] Based on the values of multiple sets of boundary conditions of the planning model, the planning model is solved respectively to obtain the planning schemes corresponding to the values of each set of boundary conditions;

[0009] Using the SHAP method, a quantitative representation of the influence degree of each boundary condition and each of the constraints on the planning scheme is obtained;

[0010] Based on the quantitative representation, the values of the corresponding boundary conditions are adjusted. Based on the adjusted values of multiple sets of boundary conditions, the planning model is solved respectively, and the planning scheme with the lowest planning cost in the solution results is selected as the optimal planning scheme.

[0011] Furthermore, each unit includes a wind power unit, a photovoltaic unit and a energy storage unit; the construction of the target function based on the load demand of the power system and the output of each unit with the lowest supply-demand mismatch degree as the goal includes:

[0012] When the sum of the wind power output and the photovoltaic power output is greater than the load demand, a supply-demand mismatch power calculation formula is constructed based on the wind power output, photovoltaic power output, load demand and energy storage charging power;

[0013] When the sum of the wind power output and the photovoltaic power output is not greater than the load demand, a supply-demand mismatch power calculation formula is constructed based on the wind power output, photovoltaic power output, load demand and energy storage discharging power;

[0014] Based on the supply-demand mismatch power calculation formula for each time period and the load demand for each time period, a target function is constructed with the lowest supply-demand mismatch degree as the goal.

[0015] Furthermore, the boundary conditions of the planning model include:

[0016] Load demand, wind power capacity factor, photovoltaic capacity factor, maximum carbon emission, carbon emission intensity of thermal power units, maximum upward / downward flexible regulation capacity of thermal power units, power upper limit of thermal power units, power upper limit of transmission lines, phase angle of transmission nodes, charging / discharging efficiency of long-term energy storage, charging / discharging efficiency of short-term energy storage.

[0017] Furthermore, using the SHAP method to obtain a quantitative representation of the influence degree of each boundary condition and each of the constraints on the planning scheme includes:

[0018] Conduct a global interpretability analysis on the planning model, and calculate the marginal contribution SHAP values of each boundary condition as the quantitative representation of each boundary condition;

[0019] Perform a local interpretability analysis on the planning model, and calculate the comprehensive relaxation degree value of each of the constraints as the quantitative representation of each of the constraints.

[0020] Furthermore, use the following formula to calculate the comprehensive relaxation degree value of each constraint condition:

[0021]

[0022] is the relaxation degree of constraint condition i at time t with different time resolutions; Ω cons is the set of model constraints; is the maximum output of the thermal power unit; is the output of the thermal power unit at time t in the optimal solution of the model; is the comprehensive relaxation degree of constraint condition i during the evolution of the plan towards the optimal solution; Ω dual is the set of constraint dual variables; α * , β * , γ * are the dual variables corresponding to each constraint in the local interpretability model;

[0023] The local interpretability model of the planning model is expressed as:

[0024]

[0025] x and y are optimization variables. The optimization variables in x are 0 / 1 variables, i.e., the operating status of the thermal power unit; the optimization variables in y are the solution variables of the model; c is the coefficient column vector corresponding to the objective function; D, F, G, and K are the coefficient matrices of the variables under the corresponding constraint conditions; d and h are constant column vectors; Dy ≤ d represents the inequality constraint in the planning model; Ky = 0 is the equality constraint in the planning model; Fx + Gy ≥ h corresponds to the constraint condition containing 0 / 1 variables in the planning model.

[0026] Furthermore, adjust the value of its corresponding boundary condition based on the quantitative representation, solve the planning model respectively based on multiple sets of boundary condition values obtained after adjustment, and select the planning scheme with the lowest planning cost in the solution results as the optimal planning scheme, including:

[0027] Sort the SHAP values from high to low, and sort the comprehensive relaxation degree values from low to high respectively;

[0028] Respectively take the boundary conditions corresponding to multiple of the top-ranked SHAP values and adjust their corresponding values within their value ranges, and / or take the boundary conditions included in the constraints corresponding to multiple of the top-ranked comprehensive relaxation degree values and adjust the corresponding values within their value ranges to obtain multiple sets of boundary condition values, and solve the model respectively based on these multiple sets of boundary condition values;

[0029] Based on the multiple obtained solution results, calculate the corresponding planning costs respectively;

[0030] Select the planning scheme corresponding to the lowest planning cost as the optimal planning scheme.

[0031] Further, the objective function is expressed as:

[0032]

[0033] where α mis is the system supply-demand mismatch degree, ΔM t is the system supply-demand mismatch power, are the wind power and photovoltaic power outputs at time t respectively, is the load demand at time t, are the charging and discharging powers of the energy storage respectively.

[0034] Further, the carbon emission constraint is expressed as:

[0035]

[0036] where, represents the set of thermal power units in region k; T represents the total duration; is the output of thermal power unit g at time t; γ car is the carbon emission intensity of the thermal power unit; Δt is the time interval; is the maximum allowable carbon emission.

[0037] Further, the power balance constraint is expressed as:

[0038]

[0039] where the subscript k represents region k; respectively represent the set of thermal power units, new energy units, load nodes and energy storage units in region k; represents the set of tie lines between region k and other regions; is the output of thermal power unit g at time t; are the outputs of wind turbine s and photovoltaic unit s′ at time t respectively; and They are the total discharge power and total charge power of energy storage in area k respectively; is the transmission power of tie line l between area k and other areas at time t; represents the load demand of area k at time t; and represent the discharge power and charge power of short-term energy storage n at time t respectively, and represent the discharge power and charge power of long-term energy storage m at time t respectively.

[0040] On the other hand, the present invention also provides a multi-resource collaborative planning system for a power system based on attribution analysis, including:

[0041] A planning model management module, which is used to construct an objective function with the lowest supply-demand mismatch degree based on the load demand of the power system and the output of each unit, and construct a planning model based on power balance constraints, line power flow constraints, output constraints of each unit, low-carbon constraints and the objective function;

[0042] A planning model solving module, which is used to solve the planning model based on multiple sets of boundary condition values of the planning model respectively to obtain planning schemes corresponding to each set of boundary condition values;

[0043] An attribution analysis module, which is used to use the SHAP method to analyze and obtain a quantitative representation of the influence degree of each boundary condition and each of the constraints on the planning scheme;

[0044] A planning scheme formulation module, which is used to adjust the boundary condition values corresponding to it based on the quantitative representation, solve the planning model respectively based on the adjusted multiple sets of boundary condition values, and select the planning scheme with the lowest planning cost in the solution results as the optimal planning scheme.

[0045] The present invention can at least achieve one of the following beneficial effects:

[0046] By determining the influence degree of each boundary condition or constraint condition on the planning scheme through attribution analysis, adjusting the boundary conditions that have a great influence on the planning scheme or the boundary conditions corresponding to the constraints that have a great influence on the planning decision according to the different influence degrees, and determining the optimal power system planning scheme based on the adjusted boundary conditions. It can scientifically formulate optimal power system planning schemes suitable for different regions and different planning stages, and save construction costs and improve economic benefits by selecting the planning scheme with the lowest planning cost.

[0047] By using the SHAP method to analyze and obtain the quantitative representation of the influence degree of each boundary condition or constraint condition on the planning scheme respectively, the boundary conditions with great influence on the planning scheme can be determined, which can play a guiding role in practical applications. By improving the technology, the boundary conditions can be adjusted. For example, when it is found that the energy storage capacity in the boundary conditions has a great influence on the planning scheme, in the current planning, the energy storage equipment with a large energy storage capacity is selected to reduce the planning cost. At the same time, in the future system construction, the development of energy storage technology should be promoted and the energy storage capacity should be expanded.

[0048] Other features and advantages of the present invention will be described in the following specification. Moreover, some advantages can be made obvious from the specification, or can be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained from the content specifically pointed out in the specification, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The drawings are only for the purpose of showing specific embodiments, and are not considered as a limitation to the present invention. Throughout the drawings, the same reference numerals represent the same components.

[0050] Figure 1 It is a flowchart of the multi-resource collaborative planning method for the power system of the present invention;

[0051] Figure 2 It is the attribution analysis result of the boundary conditions in the method embodiment of the present invention;

[0052] Figure 3 It is the attribution analysis result of the constraint conditions in the method embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] The following will specifically describe the preferred embodiments of the present invention with reference to the drawings. The drawings form a part of this application and are used together with the embodiments of the present invention to explain the principles of the present invention, and are not used to limit the scope of the present invention.

[0054] Method Embodiment

[0055] An embodiment of the present invention discloses a multi-resource collaborative planning method for a power system based on attribution analysis, specifically including the following steps:

[0056] Step S01: Based on the load demand of the power system and the output of each unit, a target function is constructed with the lowest supply-demand mismatch degree as the goal. Based on the power balance constraint, line power flow constraint, output constraint of each unit, low-carbon constraint and the target function, a planning model is constructed;

[0057] Step S02: Based on the multiple sets of boundary condition values of the planning model, the planning model is solved respectively to obtain the planning schemes corresponding to each set of boundary condition values;

[0058] Step S03, using the SHAP method to analyze and obtain a quantitative representation of the degree of influence of each boundary condition and each constraint on the planning scheme;

[0059] Step S04: adjusting the corresponding boundary condition values ​​based on the quantitative representation, solving the planning model based on the multiple groups of boundary condition values ​​obtained after the adjustment, and selecting the planning scheme with the lowest planning cost in the solution results as the optimal planning scheme.

[0060] The quantitative representation includes the marginal contribution SHAP value of each boundary condition and the comprehensive relaxation value of each constraint. Adjusting the corresponding boundary condition value based on the quantitative representation includes: sorting the SHAP values ​​from high to low, and sorting the comprehensive relaxation values ​​from low to high; taking the boundary conditions corresponding to the top-ranked SHAP values ​​and adjusting their corresponding values, and / or taking the constraint conditions corresponding to the top-ranked comprehensive relaxation values ​​and adjusting the corresponding values ​​of the boundary conditions included in these constraint conditions.

[0061] This embodiment discloses a method for coordinated planning of multiple resources of a power system based on attribution analysis, which determines the degree of influence of each boundary condition or constraint condition on the planning scheme through attribution analysis, adjusts the boundary conditions corresponding to the boundary conditions that have a great influence on the planning scheme or the constraint conditions that have a great influence on the planning decision according to the different degrees of influence, and determines the optimal power system planning scheme based on the adjusted boundary conditions. The method of this embodiment can scientifically formulate the optimal power system planning scheme suitable for different regions and different planning stages, and save construction costs and improve economic benefits by selecting the planning scheme with the lowest planning cost.

[0062] Specifically, in step S01, the power system refers to a power generation system composed of multiple generator sets. Each unit includes wind power, photovoltaic, thermal power and energy storage units; the objective function constructed based on the load demand of the power system and the output of each unit with the goal of minimizing the mismatch between supply and demand includes:

[0063] When the sum of wind power output and photovoltaic output is greater than the load demand, a supply-demand mismatch power calculation formula is constructed based on wind power output, photovoltaic output, load demand and energy storage charging power;

[0064] When the sum of wind power output and photovoltaic output is not greater than the load demand, a supply-demand mismatch power calculation formula is constructed based on wind power output, photovoltaic output, load demand and energy storage discharge power;

[0065] Based on the power calculation formula of supply-demand mismatch in each time period and the load demand in each time period, the objective function is constructed with the goal of minimizing the mismatch between supply and demand.

[0066] Furthermore, the objective function is expressed as:

[0067]

[0068] Among them, α mis is the system supply-demand mismatch degree, and ΔM t is the system supply-demand mismatch power. are the wind power and PV power outputs at time t respectively, is the load demand at time t, are the charging and discharging powers of the energy storage respectively.

[0069] Specifically, in step S01, the power balance constraint is expressed as:

[0070]

[0071] Among them, the subscript k represents area k; represent the sets of thermal power units, new energy units (wind turbines and PV units), load nodes, and energy storage units in area k respectively; represents the set of tie lines between area k and other areas; is the output of thermal power unit g at time t; are the outputs of wind turbine s and PV unit s' at time t respectively; and are the total discharging power and total charging power of the energy storage in area k respectively; is the transmission power of tie line l between area k and other areas at time t; represents the load demand in area k at time t; and are the discharging power and charging power of short-term energy storage n at time t respectively, and are the discharging power and charging power of long-term energy storage m at time t respectively.

[0072] Specifically, in step S01, the line power flow constraint is expressed as:

[0073]

[0074] Among them, is the transmission power of tie line l between area k and other areas at time t; θ a,t and θ b,t are the phase angles of the nodes connected to line l respectively, and x L is the reactance of line l, is the upper limit of the line transmission power.

[0075] Specifically, in step S01, the output constraints of each unit include new energy output constraints, thermal power output constraints, and energy storage operation constraints.

[0076] Furthermore, the new energy output constraint is expressed as:

[0077]

[0078] Wherein, and are the installed capacities of the wind turbine and photovoltaic units respectively, and are the wind power capacity factor and photovoltaic capacity factor respectively.

[0079] Furthermore, the thermal power output constraint is expressed as:

[0080]

[0081] Wherein, is the output of thermal power unit g at time t; and are the upper and lower limits of the output of thermal power unit g respectively; ν g,t is a 0 / 1 variable representing the operating state of the thermal power unit, 1 indicates the thermal power unit is operating, and 0 indicates the thermal power unit is shut down; and are the maximum upward flexible regulation capacity and maximum downward flexible regulation capacity of thermal power unit h at time t respectively.

[0082] Furthermore, the energy storage operation constraints include short-term energy storage constraints and long-term energy storage constraints.

[0083] The short-term energy storage constraint is expressed as:

[0084]

[0085] Wherein, and are the maximum capacity and maximum power of the short-term energy storage respectively, with subscripts ch and ds representing charging and discharging, is the capacity of the short-term energy storage at time t, and are the capacities of the short-term energy storage at the initial stage and end of operation respectively, η ch and η ds are the charging efficiency and discharging efficiency of the short-term energy storage respectively, and are the charging and discharging powers of the short-term energy storage at time t respectively. and are the maximum charging and maximum discharging powers of the short-term energy storage respectively.

[0086] The long-term energy storage constraint is expressed as:

[0087]

[0088] Among them, and are the capacity, charging and discharging power of long-term energy storage at time t respectively; η LD,ds and η LD,ch are the discharging efficiency and charging efficiency of long-term energy storage; Δt is the time interval; and are the capacities of the long-term energy storage at the initial stage and the end of operation respectively; and are the upper and lower limits of the capacity-power ratio of long-term energy storage respectively; are the maximum capacity, maximum discharging power, and maximum charging power of long-term energy storage respectively.

[0089] Specifically, in step S01, the low-carbon constraint is expressed as:

[0090]

[0091] Among them, γ car is the carbon emission intensity of the thermal power unit; Δt is the time interval; T is the total duration; is the maximum allowable carbon emission.

[0092] Specifically, in step S01, a planning model is constructed based on power balance constraints, line power flow constraints, output constraints of each unit, low-carbon constraints, and the objective function.

[0093] Furthermore, the boundary conditions of the planning model include: load demand wind power capacity factor photovoltaic capacity factor maximum carbon emission carbon emission intensity γ of the thermal power unit car 、maximum upward / downward flexible regulation capacity of the thermal power unit and power upper limit of the thermal power unit line transmission power upper limit transmission node phase angle θ a,t and θ b,t 、charging / discharging efficiency of long-term energy storage, charging / discharging efficiency of short-term energy storage.

[0094] Specifically, in step S02, based on the set values of multiple groups of boundary conditions, the planning model is solved respectively to obtain the planning schemes corresponding to the values of each group of boundary conditions. The planning schemes include: planning results of multi-time scale energy storage (maximum capacity of long-term energy storage), (maximum charging power of long-term energy storage), (Maximum discharge power of long-term energy storage), (Maximum capacity of short-term energy storage) and (Maximum power of short-term energy storage), as well as the installed capacity P of new wind and solar energy units PV,inv and P W,inv , and the operating conditions of each unit (Output of thermal power unit g at time t), (Output of wind turbine unit s at time t), (Output of photovoltaic unit s' at time t), (Discharge power of short-term energy storage n at time t), (Charging power of short-term energy storage n at time t), (Discharge power of long-term energy storage m at time t), (Charging power of long-term energy storage m at time t), as well as the power mismatch ΔM between system supply and demand t .

[0095] Specifically, in step S03, based on the multiple planning schemes obtained in step S02, the SHAP method is used to analyze and obtain a quantitative representation of the influence degree of each boundary condition or constraint condition on the planning scheme, including:

[0096] Conduct a global interpretability analysis on the planning model, and calculate the marginal contribution SHAP value of each boundary condition as the quantitative representation of each boundary condition;

[0097] Conduct a local interpretability analysis on the planning model, and calculate the comprehensive relaxation degree value of each constraint condition as the quantitative representation of each constraint condition.

[0098] Furthermore, for the global interpretability analysis of the planning model, under all possible input combinations of boundary conditions (features), the Shapley Value of each boundary condition (feature) is calculated and weighted averaged, and SHAP is used to achieve the global interpretation of the boundary conditions (features), and the interaction coupling effect between different boundary condition (feature) parameters is captured. The calculation process is as follows:

[0099]

[0100] Among them, g(x) represents calculating the SHAP value of the planning model; x i ∈{0,1} I , indicating whether a specific input boundary condition has an impact on the output planning result; φ i is the SHAP value representing the contribution degree of variable i to the planning result g of the planning model; φ0 = E(g(x)) is the reference value of a specific input boundary.

[0101] Among them, the marginal contribution SHAP value φ of each boundary condition i is calculated as follows:

[0102]

[0103] where M is the set of boundary conditions (feature inputs), I = |M| is the number of feature inputs, and S represents a subset of feature inputs. The importance of each input boundary is calculated by computing the difference in importance when the input boundary is present and absent in the planning scheme. g S∪{i} (x S∪{i} ) represents the model output when the i-th feature input appears; g S (x S ) represents the model output when the i-th factor does not appear; is the weighted average of all possible subsets S in M. If φ j (g, x) > 0, it indicates the promoting effect of the boundary feature input i on improving the planning benefit, thereby inferring the dominant factors corresponding to the planning schemes under different boundaries.

[0104] For example Figure 2 is the attribution analysis result of the boundary condition, and the figure shows the marginal contribution SHAP values of each boundary condition.

[0105] Furthermore, perform local interpretability analysis on the planning model, and use the comprehensive relaxation degree value of each constraint condition to reflect the looseness or tightness of a single model constraint condition under specific feature inputs.

[0106] Specifically, the compact form of the planning model can be expressed as follows:

[0107]

[0108] where x and y are optimization variables. The optimization variables in x are 0 / 1 variables, i.e., the operating status of thermal power units; the optimization variables in y are the solution variables of the model; c is the coefficient column vector corresponding to the objective function; D, F, G, and K are the coefficient matrices of variables under corresponding constraint conditions; d and h are constant column vectors; Dy ≤ d represents the inequality constraint in the planning model; Ky = 0 is the equality constraint in the planning model; Fx + Gy ≥ h corresponds to the constraint condition containing 0 / 1 variables in the planning model; α * , β * , γ * are the dual variables corresponding to each constraint in the local interpretability model.

[0109] The expressions corresponding to x and y are:

[0110]

[0111] It should be noted that the dual variable of a single constraint, that is, the slack, varies with the time resolution of the constraint. To measure the overall slack degree of a single constraint in the planning model, when the constraint tightens, the slack will be closer to 0. At this time, the value of the dual variable is not 0, but when the constraint relaxes, the value of the dual variable is 0.

[0112] Furthermore, the calculation method of the comprehensive slack of each constraint condition is: sum the absolute values of the dual variables at each resolution, which reflects the tightness of a specific constraint in the planning model. For example, after inputting the parameters of specific boundary conditions, the optimal operating conditions and planning schemes of each unit are solved, and the distances from the upper and lower bounds of the constraint are calculated.

[0113] Furthermore, the comprehensive slack value of each constraint condition is calculated using the following formula:

[0114]

[0115] is the slack of constraint condition i at time t under different time resolutions; Ω cons is the set of model constraints; is the maximum output of the thermal power unit; is the output of the thermal power unit at time t in the optimal solution of the model; is the comprehensive slack of constraint condition i during the evolution of the plan towards the optimal solution; Ω dual is the set of constraint dual variables. By calculating the comprehensive slack of each constraint condition, the contribution of each constraint to the optimal solution is measured. Constraints with low dual variable values have higher comprehensive slack, indicating that the constraint conditions have a large margin in the current optimal solution and a small marginal contribution to the planning goal; constraints with high dual variable values have lower comprehensive slack, indicating that the constraint is a tight constraint or close to a tight constraint and has a greater marginal contribution to the planning goal, thereby identifying the tight constraints that make important contributions to the planning results. As Figure 3 described is the attribution analysis result of the constraint conditions, and the figure shows the comprehensive slack values of each constraint condition.

[0116] Specifically, in step S04, based on the quantization representation, the corresponding boundary condition values are adjusted, and the planning model is solved respectively based on the adjusted multiple sets of boundary condition values, and the planning scheme with the lowest planning cost in the solution results is selected as the optimal planning scheme, including:

[0117] Sort the SHAP values from high to low and sort the comprehensive slack values from low to high respectively;

[0118] Respectively take the boundary conditions corresponding to multiple of the top-ranked SHAP values and adjust their corresponding values within their value ranges, and / or take the boundary conditions included in the constraints corresponding to multiple of the top-ranked comprehensive relaxation values and adjust the corresponding values within their value ranges to obtain multiple sets of boundary condition values, and solve the model based on these multiple sets of boundary condition values;

[0119] Based on the multiple obtained solution results, calculate the corresponding planning costs respectively;

[0120] Select the planning scheme corresponding to the lowest planning cost as the optimal planning scheme.

[0121] Exemplarily, such as Figure 2 shown, the boundary conditions that contribute more to the planning model include load demand, maximum carbon emissions, upward regulation capacity of thermal power units, long-term energy storage discharge efficiency, short-term energy storage charge-discharge efficiency; as Figure 3 stated, the constraint conditions that contribute more to the planning model include power balance constraint, carbon emission constraint, long-term energy storage capacity constraint. Therefore, the boundary conditions that may need to be adjusted are determined to include the allowable maximum carbon emissions, upward regulation capacity of thermal power units, long-term energy storage discharge efficiency, short-term energy storage charge-discharge efficiency, and the boundary conditions involved in the constraint conditions such as the maximum long-term energy storage capacity, maximum discharge power, maximum charge power, etc.

[0122] It should be noted that when selecting the boundary conditions corresponding to multiple of the top-ranked SHAP values and / or the constraint conditions corresponding to multiple of the top-ranked comprehensive relaxation values, the corresponding boundary condition values need to be adjusted in combination with the actual application situation. Exemplarily, such as Figure 2 shown, the boundary condition with the largest SHAP value contributing to the solution result of the planning model is the load demand. In actual planning, the load demand needs to meet the production and living needs. Therefore, when adjusting the load demand, it needs to be adjusted within the limit of meeting the load requirements in the planning. Similarly, such as Figure 2 , for the maximum carbon emissions that contribute more to the solution result of the planning model, the boundary condition values also need to be adjusted under the condition of meeting the policy requirements of the corresponding region and corresponding planning stage. For the boundary conditions that can be adjusted through technological improvement, such as the upward regulation capacity of thermal power units, energy storage charge-discharge efficiency, etc., the boundary condition values can be adjusted in combination with the actual achievable situation of technological improvement during actual implementation (such as through reasonable model selection).

[0123] A multi - resource collaborative planning method for power systems based on attribution analysis disclosed in this embodiment determines the influence degree of each boundary condition or constraint condition on the planning scheme through attribution analysis, adjusts the boundary conditions that have a great impact on the planning scheme or the corresponding boundary conditions of the constraints that have a great impact on the planning decision according to the different influence degrees, and determines the optimal power system planning scheme based on the adjusted boundary conditions. It can scientifically formulate the optimal power system planning scheme suitable for different regions and different planning stages, and save construction costs by selecting the planning scheme with the lowest planning cost, thus improving economic benefits.

[0124] By using the SHAP method to analyze and obtain a quantitative representation of the influence degree of each boundary condition or constraint condition on the planning scheme, the boundary conditions that have a great impact on the planning scheme can be determined, which can play a guiding role in practical applications. The adjustment of boundary conditions is realized through the improvement of technology. For example, when it is found that the energy storage capacity in the boundary conditions has a great impact on the planning scheme, in the current planning, the planning cost can be reduced by selecting energy storage devices with a large energy storage capacity, and at the same time, in the future system construction, the development of energy storage technology should be promoted to expand the energy storage capacity.

[0125] System embodiment

[0126] Another specific embodiment of the present invention discloses a multi - resource collaborative planning system for power systems based on attribution analysis, including a planning model management module, a planning model solving module, an attribution analysis module, and a planning scheme formulation module.

[0127] Among them, the planning model management module is used to construct an objective function with the lowest supply - demand mismatch degree based on the load demand of the power system and the output of each unit, and construct a planning model based on power balance constraints, line power flow constraints, output constraints of each unit, low - carbon constraints, carbon cost calculation formulas, and the objective function;

[0128] The planning model solving module is used to solve the planning model based on multiple sets of boundary condition values of the planning model to obtain the planning schemes corresponding to each set of boundary condition values;

[0129] The attribution analysis module is used to analyze and obtain a quantitative representation of the influence degree of each boundary condition or constraint condition on the planning scheme by using the SHAP method;

[0130] The planning scheme formulation module is used to adjust the corresponding boundary condition values based on the quantitative representation, solve the planning model based on the adjusted multiple sets of boundary condition values, and select the planning scheme with the lowest planning cost in the solution results as the optimal planning scheme.

[0131] Compared with the prior art, the beneficial effects provided by this embodiment are basically the same as those provided by the method embodiment, and will not be elaborated here one by one.

[0132] It should be noted that the above embodiments are based on the same inventive concept. For the parts not repeatedly described, reference can be made to each other.

[0133] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A multi-resource collaborative planning method for power systems based on attribution analysis, characterized in that, The method includes the following steps: Based on the load demand of the power system and the output of each unit, a target function is constructed with the lowest supply-demand mismatch degree as the goal. Based on the power balance constraint, line power flow constraint, output constraint of each unit, low-carbon constraint and the target function, a planning model is constructed; Based on the values of multiple sets of boundary conditions of the planning model, the planning model is solved respectively to obtain the planning schemes corresponding to the values of each set of boundary conditions; The SHAP method is used to analyze and obtain a quantitative representation of the influence degree of each boundary condition and each of the constraints on the planning scheme; Based on the quantitative representation, the values of the corresponding boundary conditions are adjusted. Based on the adjusted values of multiple sets of boundary conditions, the planning model is solved respectively, and the planning scheme with the lowest planning cost in the solution results is selected as the optimal planning scheme.

2. The planning method according to claim 1, characterized in that, Each of the units includes a wind power unit, a photovoltaic unit and a energy storage unit; constructing the target function based on the load demand of the power system and the output of each unit with the lowest supply-demand mismatch degree as the goal includes: When the sum of the wind power output and the photovoltaic power output is greater than the load demand, a supply-demand mismatch power calculation formula is constructed based on the wind power output, photovoltaic power output, load demand and energy storage charging power; When the sum of the wind power output and the photovoltaic power output is not greater than the load demand, a supply-demand mismatch power calculation formula is constructed based on the wind power output, photovoltaic power output, load demand and energy storage discharging power; Based on the supply-demand mismatch power calculation formulas for each time period and the load demand for each time period, a target function is constructed with the lowest supply-demand mismatch degree as the goal.

3. The planning method according to claim 2, characterized in that, The boundary conditions of the planning model include: Load demand, wind power capacity factor, photovoltaic capacity factor, maximum carbon emission, carbon emission intensity of thermal power units, maximum upward / downward flexible regulation capacity of thermal power units, power upper limit of thermal power units, power upper limit of transmission lines, phase angle of transmission nodes, charging / discharging efficiency of long-term energy storage, charging / discharging efficiency of short-term energy storage.

4. The planning method according to any one of claims 1-3, characterized in that, Using the SHAP method to analyze and obtain a quantitative representation of the influence degree of each boundary condition and each of the constraints on the planning scheme includes: Performing a global interpretability analysis on the planning model, and calculating the marginal contribution SHAP values of each boundary condition as the quantitative representation of each boundary condition; Performing a local interpretability analysis on the planning model, and calculating the comprehensive relaxation degree values of each of the constraints as the quantitative representation of each of the constraints.

5. The planning method according to claim 4, characterized in that, The comprehensive relaxation degree value of each constraint condition is calculated using the following formula: is the relaxation degree of constraint condition i at time t with different time resolutions; Ω cons is the set of model constraints; is the maximum output of the thermal power unit; is the output of the thermal power unit at time t in the optimal solution of the model; is the comprehensive relaxation degree of constraint condition i during the evolution from the plan to the optimal solution; Ω dual is the set of constraint dual variables; α * , β * , γ * are the dual variables corresponding to each constraint in the local interpretability model; The local interpretability model of the planning model is expressed as: x and y are optimization variables. The optimization variables in x are 0 / 1 variables, that is, the operating state of thermal power units; the optimization variables in y are the solution variables of the model; c is the coefficient column vector corresponding to the target function; D, F, G, and K are coefficient matrices of variables under corresponding constraint conditions; d and h are constant column vectors; Dy≤d represents the inequality constraint in the planning model; Ky = 0 is the equality constraint in the planning model; Fx+Gy≥h corresponds to the constraint condition containing 0 / 1 variables in the planning model.

6. The planning method according to claim 4, characterized in that, The step of adjusting the boundary condition values ​​corresponding to the quantified representation, solving the planning model based on the multiple groups of boundary condition values ​​obtained after the adjustment, and selecting the planning scheme with the lowest planning cost in the solution results as the optimal planning scheme includes: The SHAP values ​​are sorted from high to low, and the comprehensive relaxation values ​​are sorted from low to high respectively; Respectively taking the boundary conditions corresponding to the top multiple SHAP values ​​and adjusting their corresponding values ​​within their value range, and / or taking the boundary conditions included in the constraints corresponding to the top multiple comprehensive slack values ​​and adjusting their corresponding values ​​within their value range, to obtain multiple groups of boundary condition values, and solving the model based on the multiple groups of boundary condition values; Based on the multiple solution results obtained, the corresponding planning costs are calculated respectively; The planning scheme corresponding to the lowest planning cost is selected as the optimal planning scheme.

7. The planning method according to claim 2, wherein The objective function is expressed as: Among them, α mis is the system supply-demand mismatch degree, ΔM t is the system supply-demand mismatch power, are the wind power and photovoltaic power outputs at time t respectively, is the load demand at time t, are the charging and discharging powers of the energy storage respectively.

8. The planning method according to claim 7, wherein The carbon emission constraint is expressed as: Among them, represents the set of thermal power units in region k; T represents the total duration; is the output of thermal power unit g at time t; γ car is the carbon emission intensity of the thermal power unit; Δt is the time interval; is the maximum allowable carbon emission.

9. The planning method according to claim 7, wherein The power balance constraint is expressed as: Among them, the subscript k represents the k region; respectively represent the set of thermal power units, the set of new energy units, the set of load nodes, and the set of energy storage units in region k; represents the set of tie lines between region k and other regions; is the output of thermal power unit g at time t; are respectively the outputs of wind turbine s and photovoltaic unit s′ at time t; and are respectively the total discharge power and total charge power of energy storage in region k; is the transmission power of tie line l between region k and other regions at time t; represents the load demand in region k at time t; and are respectively the discharge power and charge power of short-term energy storage n at time t, and are respectively the discharge power and charge power of long-term energy storage m at time t.

10. A multi-resource collaborative planning system for power systems based on attribution analysis, characterized in that, include: A planning model management module is used to construct an objective function based on the load demand of the power system and the output of each unit with the goal of minimizing the mismatch between supply and demand, and to construct a planning model based on power balance constraints, line flow constraints, output constraints of each unit, low-carbon constraints and the objective function; A planning model solving module, used to solve the planning model based on multiple groups of boundary condition values ​​of the planning model to obtain planning solutions corresponding to each group of boundary condition values; An attribution analysis module is used to analyze and obtain a quantitative representation of the influence of each boundary condition and each constraint on the planning scheme using the SHAP method; The planning scheme formulation module is used to adjust the corresponding boundary condition values ​​based on the quantitative representation, solve the planning model based on the multiple groups of boundary condition values ​​obtained after the adjustment, and select the planning scheme with the lowest planning cost in the solution results as the optimal planning scheme.

Images