A method and device for planning power supply structure considering demand-side response

Abstract

The present invention discloses a power structure planning method and device considering demand-side response. After the original load curve is generated by acquiring historical load data, the original load curve is optimized to a demand response load curve that meets the demand response according to the demand response. After the working modes of various power sources are obtained through the demand response load curve and a preset objective function, the operating cost is obtained by simulating production according to the working modes of different power sources. In this way, the power generation space of different power sources can be reasonably controlled according to the demand response and the working characteristics of different power sources, so that different power sources such as new energy power sources, thermal power sources, etc. can achieve the optimal power utilization effect.

Description

Technical Field

[0001] The present invention relates to the technical field of power supply structure planning, and in particular to a power supply structure planning method and device taking demand-side response into consideration. Background Art

[0002] Vigorously developing new energy has become a common consensus among countries around the world, and a reasonable power structure is the cornerstone for ensuring efficient consumption of new energy and safe and reliable power supply. In traditional power planning methods, the main problem faced is that the installed capacity cannot meet the power demand. In long-term power planning, with the development of the economy and society, the power demand tends to be saturated, and the main problem of the power structure has changed to the problem of insufficient flexibility of the power system.

[0003] And with the continuous development of new energy, new energy also occupies an important position in the power structure. However, due to the uncertainty of new energy output, it is impossible to reasonably regulate the power generation and power consumption of new energy. This affects the power generation space of various types of power sources and makes it impossible to effectively regulate various types of power sources. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a method and device for planning a power supply structure taking into account the demand side response, so as to improve the reasonable control of the output of different power supplies.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A power structure planning method considering demand side response comprises the steps of:

[0007] A power structure planning method considering demand side response comprises the steps of:

[0008] Obtain the historical load data line to obtain the original load curve;

[0009] Acquiring load demand characteristics, and converting the original load curve into a demand response load curve according to the load demand characteristics;

[0010] According to the demand response load curve and the preset objective function, the working mode of each type of power supply is obtained;

[0011] Calculate the unit power generation cost of each type of power source according to the working mode of each type of power source;

[0012] The working modes of various power sources are optimized according to the unit power generation cost to obtain the optimal total power generation cost and the optimal working modes of various power sources.

[0013] In order to solve the above technical problems, another technical solution adopted by the present invention is:

[0014] A power structure planning device that takes demand-side response into consideration includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, each step in the above-mentioned power structure planning method that takes demand-side response into consideration is implemented.

[0015] The beneficial effects of the present invention are as follows: after the original load curve is generated by acquiring historical load data, the original load curve is converted into a demand response load curve according to the load demand characteristics, and then the working modes of various power sources are obtained through the demand response load curve and a preset objective function. Then, according to the working modes of different power sources, simulated production is performed to obtain the operating cost, so that the power generation space of different power sources can be reasonably controlled according to the demand response and the working characteristics of different power sources, so that different power sources such as new energy power sources, thermal power sources, etc. can achieve the optimal power utilization effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic flow chart of a method for planning a power structure taking into account demand-side response according to an embodiment of the present invention;

[0017] Figure 2 A demand side response regulation effect diagram of a power structure planning method considering demand side response according to an embodiment of the present invention;

[0018] Figure 3 A seasonal regulation energy storage working effect diagram of a power structure planning method considering demand side response in an embodiment of the present invention;

[0019] Figure 4 A conventional power supply production simulation diagram of a power supply structure planning method considering demand-side response in an embodiment of the present invention;

[0020] Figure 5 A schematic diagram of the structure of a power structure planning device considering demand-side response according to an embodiment of the present invention;

[0021] Figure 6 A schematic diagram of a double-layer power supply planning model structure of a power supply structure planning method considering demand-side response according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in combination with the implementation modes and the accompanying drawings.

[0023] Please refer to Figure 1 , a power structure planning method considering demand side response, comprising the steps of:

[0024] Obtain the historical load data line to obtain the original load curve;

[0025] Acquiring load demand characteristics, and converting the original load curve into a demand response load curve according to the load demand characteristics;

[0026] According to the demand response load curve and the preset objective function, the working mode of each type of power supply is obtained;

[0027] Calculate the unit power generation cost of each type of power source according to the working mode of each type of power source;

[0028] The working modes of various power sources are optimized according to the unit power generation cost to obtain the optimal total power generation cost and the optimal working modes of various power sources.

[0029] From the above description, it can be seen that the beneficial effects of the present invention are: after generating the original load curve by acquiring historical load data, the original load curve is converted into a demand response load curve according to the load demand characteristics, and then the working modes of various power sources are obtained through the demand response load curve and the preset objective function. Then, according to the working modes of different power sources, simulated production is performed to obtain the operating cost, so that the power generation space of different power sources can be reasonably controlled according to the demand response and the working characteristics of different power sources, so that different power sources such as new energy power sources and thermal power sources can achieve the optimal power utilization effect.

[0030] Furthermore, the obtaining of the historical load data line to obtain the original load curve includes:

[0031] Obtain historical output data and load growth factor;

[0032] The original load curve is calculated based on the historical output data and the load growth coefficient.

[0033] From the above description, it can be seen that by obtaining historical output data and load growth coefficients, and then multiplying the historical output data by the load growth coefficient to obtain the original load curve, it is possible to predict the annual load time series and the new energy output time series based on the historical load, and use this to plan the annual load time series.

[0034] Furthermore, the historical output data includes a new energy output time series;

[0035] The new energy output time series includes a wind power output time series and a photovoltaic power output time series;

[0036] Obtain predicted wind power output values ​​of different partitions, and add them up to obtain the wind power output time series;

[0037] The historical photovoltaic output data and the capacity coefficient corresponding to the photovoltaic power source are obtained, and the photovoltaic output time series is obtained by multiplying them.

[0038] It can be seen from the above description that by obtaining the predicted wind power output values ​​of different partitions, adding them to obtain the wind power output time series, obtaining the historical photovoltaic output data and the capacity coefficient corresponding to the photovoltaic power source, and multiplying them to obtain the photovoltaic output time series, that is, calculating the wind power output based on spatial complementarity and calculating the photovoltaic output time series based on historical data and the capacity coefficient, the accuracy of the calculation results of the wind power output time series and the photovoltaic output time series is improved;

[0039] Further, the converting the original load curve into a demand response load curve according to the load demand characteristics includes:

[0040] Controlling the output of new energy and seasonal regulation energy storage according to the load demand characteristics and transfer function;

[0041]

[0042] Among them, P sea sto (t) is the seasonal regulation energy storage power, a positive value indicates that it is working in the power generation state, and a negative value indicates that it is working in the energy storage state; S sea T is the rated capacity of seasonal regulation energy storage; va For the power shortage season, T ab It is the season of electricity surplus. P NE va , P NE ab They are the new energy output setting values ​​for energy storage to start generating electricity in the power shortage season and the power surplus season, respectively. The renewable energy output setting value for energy storage to begin during the electricity surplus season.

[0043] From the above description, it can be seen that the output of renewable energy is controlled according to the load demand characteristics and the transfer function, that is, the energy storage and load demand can be adjusted according to the season, the electricity in the season with abundant renewable energy resources is stored in the season with scarce resources, and no electricity is stored during the power shortage period. When the output of renewable energy is less than the set value, power generation is started, and in the power surplus period, when the output of renewable energy is greater than the set value, power storage is started, and when the output of renewable energy is less than the set value, power generation is started, so as to effectively control the time series of renewable energy output.

[0044] Furthermore, after converting the original load curve into a demand response load curve according to the load demand characteristics, the method further includes:

[0045] Obtaining a net load time series according to the demand response load curve and the new energy output time series;

[0046] The net load time series is converted by the peak shaving and valley filling function to obtain the non-renewable energy output time series.

[0047] From the above description, it can be seen that by subtracting the new energy output time series from the demand response load curve to obtain the net load time series, and then adjusting the energy storage daily to smooth the peaks and fill the valleys of the net load time series, it is possible to obtain the load time series after considering the energy storage and new energy output, and distinguish the time series of new energy from non-new energy, so that the new energy output and non-new energy output can be controlled separately.

[0048] Furthermore, calculating the unit power generation cost of each type of power supply according to the working mode of each type of power supply includes:

[0049] Performing a non-sequential random production simulation according to preset constraints and the working modes of the different power supplies to obtain the operating cost;

[0050] The preset constraints include power balance constraints, carbon emission constraints, system reliability constraints and buildable capacity constraints.

[0051] From the above description, it can be seen that by setting power balance constraints, carbon emission constraints, system reliability constraints and buildable capacity constraints, and using non-sequential random production simulation, the reliability index and economic index of the system can be accurately calculated.

[0052] Furthermore, the working modes of various power sources obtained according to the demand response load curve and the preset objective function include:

[0053] Determine whether the current state is the initial state, and if so, generate a set of installed capacity data satisfying the constraint conditions according to the constraint conditions;

[0054] If not, generating a set of optimal installed capacity data according to the objective function;

[0055] The objective function is to minimize the total annual operating cost, that is:

[0056] where c i is the unit power generation cost of the i-th power source, E i Its annual power generation.

[0057] From the above description, it can be seen that when the initial installation scale is selected, it is only necessary to give a set of installation capacity data arbitrarily under the premise of satisfying the constraints. When it is not the first installation, the simulation iterative optimization is carried out according to the objective function - annual total operating cost to obtain the final installation plan.

[0058] Furthermore, the operating costs obtained by simulating production according to the working modes of the various types of power supplies include:

[0059] Calculate the unit power generation cost of each type of power source according to the working mode of each type of power source;

[0060] c 清洁 =c1+c2·T / t;

[0061]

[0062] c 储能 =c1;

[0063] Among them, c1, c2, and c3 correspond to fixed costs, variable power generation costs, and excess carbon emission costs, respectively; T is the planned utilization hours, and t is the actual utilization hours.

[0064] From the above description, it can be seen that the total power generation cost of the system is calculated according to the working modes of different power sources obtained and combined with the power generation costs of different power sources, so that the working modes of different power sources can be further adjusted from the perspective of power generation cost to achieve the optimal power generation cost.

[0065] Furthermore, converting the original load curve into a demand response load curve according to the load demand characteristics includes:

[0066] The original load curve is converted into a peak-cutting one by using the peak-cutting function:

[0067]

[0068] And get the total peak shaving power E for the day DR peak ; Among them, P DR (t) is the response scale at time i, t max is the maximum load moment, P DRmax peak is the maximum load moment response scale, [t b ,t f ] is the response period.

[0069] The original load curve is converted into a valley-filling curve by using the valley-filling function:

[0070]

[0071] From the above description, it can be seen that by using the peak shaving function to transform the original load curve into a peak-peak load, the load in the peak power consumption period can be transferred to the valley power consumption period, reducing the maximum load that the system needs to bear. By using the valley filling function to transform the original load curve into a valley-filling function, the load in the peak power consumption period can be transferred to the valley power consumption period, avoiding the system load being too low in the valley power consumption period, which reduces the system's power generation and power consumption efficiency, makes the system load more uniform, and improves the system stability.

[0072] Another embodiment of the present invention provides a power structure planning device that takes demand-side response into consideration, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, each step in a power structure planning method that takes demand-side response into consideration is implemented as described above.

[0073] The above-mentioned power structure planning method and device considering demand-side response of the present invention can be applied to the power structure planning of dynamic power generation cost, and plan and control different power sources such as new energy, thermal power and nuclear power. The following is an explanation through specific implementation methods:

[0074] Please refer to Figure 1 , a power structure planning method considering demand side response, comprising the steps of:

[0075] S1. Obtain the historical load data line to obtain the original load curve;

[0076] S11, obtaining historical output data and load growth coefficient;

[0077] S12, calculating the original load curve according to the historical output data and the load growth coefficient;

[0078] The historical output data includes the new energy output time series and the annual load time series; the new energy output time series includes the wind output time series and the photovoltaic output time series; the wind output time series is obtained by obtaining the predicted wind power output values ​​of different partitions and adding them together; the historical photovoltaic output data and the capacity coefficient corresponding to the photovoltaic power source are obtained and multiplied to obtain the photovoltaic output time series;

[0079] S2. Obtaining load demand characteristics, and converting the original load curve into a demand response load curve according to the load demand characteristics;

[0080] Please refer to Figure 2 Specifically, in an optional implementation, considering the demand side response effect in different time periods every day, a quadratic function is used to construct a demand side response process model, and the original load curve is subjected to daily regulation and energy storage control to obtain the load time series after the demand side response The response method is to keep the daily power consumption unchanged and move the load of the peak hours in the afternoon and evening to the low-load period at night. The specific response process is as follows:

[0081] The original load curve is converted into a peak-cutting one by using the peak-cutting function:

[0082]

[0083] And get the total peak shaving power E for the day DRpeak ; Among them, P DR (t) is the response scale at time i, t max is the maximum load moment, P DRmax peak is the maximum load moment response scale, [t b ,t f ] is the response period; the noon peak period and evening peak period can be selected according to the typical daily load characteristics. The response scale is the largest at the time of maximum load, and the response scale gradually decreases at other times;

[0084] The original load curve is converted into a valley-filling curve by using the valley-filling function:

[0085]

[0086] In the valley filling response process, the response scale is the largest at the lowest load, and the response scale gradually decreases at other response times. The valley filling period can be selected at night according to the typical daily load characteristic curve;

[0087] In another optional implementation, considering the different demand-side response effects in each quarter, seasonal regulation energy storage is used to smooth the peaks and fill the valleys of the new energy output time series, and according to the load demand, the electricity in the season with abundant new energy resources is stored in the season with scarce resources; the main function of seasonal regulation energy storage is the seasonal transfer of electricity; since the output of new energy has obvious seasonal characteristics, the use of seasonal regulation energy storage can transfer the surplus electricity in the surplus season to the deficit season. Considering that seasonal energy storage is mainly used to solve the long-term fluctuation (week, month, season) of new energy output, its electricity mainly comes from new energy, so the work of seasonal energy storage can be regarded as smoothing the peaks and filling the valleys of the new energy output curve. The main modeling principles include:

[0088] (1) Power generation E gen Less than the storage capacity E sto Multiply by the conversion efficiency η;

[0089] E gen ≤ηE sto ;

[0090] (1) Seasonal regulation storage does not store electricity during power shortage periods, and starts generating electricity when the output of renewable energy is less than the set value; during power surplus periods, it starts storing electricity when the output of renewable energy is greater than the set value, and starts generating electricity when the output of renewable energy is less than the set value. Specifically:

[0091] Controlling the output of the new energy according to the demand response and the transfer function;

[0092]

[0093] Among them, P seasto (t) is the seasonal regulation energy storage power, a positive value indicates that it is working in the power generation state, and a negative value indicates that it is working in the energy storage state; S sea T is the rated capacity of seasonal regulation energy storage; va For the power shortage season, T ab It is the season of electricity surplus. P NE va , P NE ab They are the new energy output setting values ​​for energy storage to start generating electricity in the power shortage season and the power surplus season, respectively. The new energy output setting value for energy storage to start storing electricity during the electricity surplus season;

[0094] In a specific scenario, such as a southern city, the power shortage mainly occurs in summer, and the power surplus occurs in winter. The seasonal regulation energy storage working mode can be set as follows:

[0095] (1) No energy storage is performed from June to August. Energy storage is performed during the remaining periods when the output of renewable energy is greater than the annual average output level;

[0096] (2) From June to August, when the output of renewable energy is less than 50%, power generation starts; in other periods, when the output of renewable energy is less than 20%, power generation starts; the effect of seasonal regulation energy storage is as follows: Figure 3 As shown;

[0097] Secondly, consider the working characteristics of daily regulation energy storage; the main function of daily regulation energy storage is to reduce the peak of the daily load curve and fill the valley, using electricity during the peak load and generating electricity during the valley load;

[0098] After obtaining the new energy output time series and the demand response load curve,

[0099] S21. Obtaining a net load time series according to the demand response load curve and the new energy output time series;

[0100] S22, using daily regulation of energy storage to smooth the peaks and fill the valleys of the net load time series, to obtain a non-new energy output time series; that is, to obtain a load time series after considering energy storage and new energy;

[0101] S3. According to the demand response load curve and the preset objective function, the working mode of each type of power source is obtained; that is, after the non-renewable energy output time series and the new energy output time series are brought into the preset objective function, the optimal installation plan is output;

[0102] S4. Performing simulated production according to the working modes of the different power sources to obtain operating costs.

[0103] Embodiment 2

[0104] The difference between this embodiment and the first embodiment is that the process of simulated production is limited;

[0105] According to the preset constraints and the working modes of the different power supplies, a non-sequential random production simulation is performed to obtain the specific operating cost:

[0106] In an optional implementation, if the unit includes nuclear power, thermal power and hydropower, the working characteristics and working positions of each type of power source are considered, and the working order of the power source is reasonably arranged, such as the working order of the unit is nuclear power-minimum technical output of thermal power-forced output of hydropower-thermal power-hydropower; wherein the preset constraints include power balance constraints, carbon emission constraints, system reliability constraints and buildable capacity constraints; the power balance constraint is: thermal power + 0.5 hydropower + nuclear power > maximum net load; carbon emission constraint: thermal power generation < carbon emission allowable generation; system reliability constraint is: load loss probability < 0.01; buildable capacity constraint is: installed capacity of each type of power source < buildable capacity;

[0107] Specifically, the non-sequential random production simulation is a random production simulation based on an equivalent continuous load curve. The formation process is shown in Figure 4 ; In the figure, the horizontal axis x is the load level, and the vertical axis T is the duration, that is, any point (x1, T) on the curve means that the duration of the system load greater than x1 is T1;

[0108] Among them, f0(x) in the figure is the original load continuity curve of the system, that is, the power load that the system needs to bear; in the random production simulation process, the random failure and shutdown of the generator is equivalent to the correction process of the continuous load curve; let the capacity of the first generator be C1, when its failure rate is 0, the load it can bear is the shaded area in the figure; when its failure rate is 1, the load it bears should be borne by all the other units at this time;

[0109] For example, in the original state, generator 1 and other generators share the equivalent load of f0(x). When generator 1 fails, it is equivalent to generator 1 and other generators sharing the equivalent load of f0(x-C1). Assuming that the failure probability of the generator is q, the actual power generation of the system is C1T(1-q). After considering the probability of random outage of generator 1, the equivalent continuous load curve of the system will be between f0(x) and f0(x-C1), that is, it becomes:

[0110] f1(x)=(1-q1)f0(x)+q1f0(x-C1);

[0111] Repeat the above process for all generator sets in sequence; as the generators are put into operation one after another, the equivalent continuous load curve of the system is adjusted accordingly; when all generator sets are put into operation, the operating cost (economic index) and load loss probability (reliability index) of the system can be calculated;

[0112] Among them, a piecewise function is used to describe the power generation cost of various types of power sources; various types of power sources are divided into three categories: clean power sources, carbon emission power sources and energy storage power sources. The power generation cost is calculated according to the working modes of different power sources as follows:

[0113] c 清洁 =c1+c2·T / t;

[0114]

[0115] c 储能 =c1;

[0116] Among them, c1, c2, and c3 correspond to fixed costs, variable power generation costs, and excess carbon emission costs, respectively;

[0117] For example, if the power supply structure of a certain region in the south is optimized under the carbon neutrality target in 2050, the optimized power supply structure and operation conditions are shown in Table 1. The total annual cost of the system is 18.412 billion yuan, and the probability of load loss is 0.13%. The power supply solution that does not consider optimization and only considers resource endowment can reduce costs by 13.4%, which can achieve better economy and planning effects.

[0118] Table 1. Considering only resource endowment scheme and operation status (total annual cost 21.33 billion yuan)

[0119]

[0120]

[0121] Table 2. Optimization scheme and operation status

[0122]

[0123] Embodiment 3

[0124] Please refer to Figure 5 , a power structure planning device considering demand side response, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, each step in a power structure planning method considering demand side response in Embodiment 1 or 2 is implemented;

[0125] Please refer to Figure 6, a two-layer power planning model is established to consider the system reliability constraints, power balance constraints and carbon emission constraints; the upper layer uses a particle swarm algorithm to optimize the capacity of various types of power sources, including wind power, photovoltaic power, nuclear power, thermal power, hydropower, daily regulation energy storage, and seasonal regulation energy storage. The objective function is to minimize the annual total operating cost, that is:

[0126]

[0127] where c i is the unit power generation cost of the i-th power source, E i is its annual power generation; the initial power installed capacity can be arbitrarily given a set of data under the constraints, and the optimal installed capacity is obtained through iterative optimization; the lower layer uses production simulation to calculate the reliability index and economic index of the system to obtain the optimal power structure of the system; the demand-side response means are incorporated into the power planning level, and the problem of various types of power sources squeezing each other's power generation space brought about by large-scale new energy access is considered, and a dynamic power generation cost modeling method for various types of power sources is proposed; considering the working characteristics of energy storage and various types of conventional power sources for production simulation, a two-layer power structure optimization planning model with construction and operation economy as the goal is established. The upper layer uses particle swarm algorithm to optimize the capacity of various types of power sources, and the lower layer uses production simulation to calculate the reliability index and economic index of the system, which solves the problem that the application potential of new technologies such as demand-side response and energy storage is not fully considered in traditional power planning, and provides an effective solution for guiding the power planning of new power systems under the dual carbon goals.

[0128] In summary, the present invention provides a power structure planning method and device that considers demand-side response, and proposes a modeling method for demand-side response. After obtaining historical load data to generate an original load curve, the original load curve is optimized to a demand response load curve that meets the demand response according to the demand response, and then the optimal working mode of different power sources is obtained in combination with the preset constraint conditions. Finally, the optimal power structure is obtained according to the working mode of different power sources, so that the power generation space of different power sources can be reasonably controlled according to the demand response and the working characteristics of different power sources, so that different power sources such as new energy power sources and thermal power sources can achieve the optimal power utilization effect.

[0129] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A power structure planning method considering demand-side response, characterized in that , including the steps: Obtain the historical load data line to obtain the original load curve; Acquiring load demand characteristics, and converting the original load curve into a demand response load curve according to the load demand characteristics; According to the demand response load curve and the preset objective function, the working mode of each type of power supply is obtained; Calculate the unit power generation cost of each type of power source according to the working mode of each type of power source; Optimize the working modes of various power sources according to the unit power generation cost to obtain the optimal total power generation cost and the optimal working modes of various power sources; The converting the original load curve into a demand response load curve according to the load demand characteristics comprises: Controlling the output of new energy and seasonal regulation energy storage according to the load demand characteristics and transfer function; Among them, P sea sto (t) is the seasonal regulation energy storage power, a positive value indicates that it is working in the power generation state, and a negative value indicates that it is working in the energy storage state; S sea T is the rated capacity of seasonal regulation energy storage; va For the power shortage season, T ab It is the season of electricity surplus. P NE va , P NE ab Source output setting value, The new energy output setting value for energy storage to start storing electricity during the electricity surplus season; The operating costs obtained by simulating production according to the working modes of the various power sources include: Calculate the power generation cost according to the working modes of the various types of power sources; c 清洁 =c1+c2·T / t; c 储能 =c1; Among them, c1, c2, and c3 correspond to fixed costs, variable power generation costs, and excess carbon emission costs, respectively; T is the planned utilization hours, and t is the actual utilization hours.

2. A method for planning a power structure considering demand-side response according to claim 1, characterized in that: The obtaining of the historical load data line to obtain the original load curve comprises: Obtain historical output data and load growth factor; The original load curve is calculated based on the historical output data and the load growth coefficient.

3. A method for planning a power structure considering demand-side response according to claim 2, characterized in that: The historical output data includes a new energy output time series; The new energy output time series includes a wind power output time series and a photovoltaic power output time series; Obtaining predicted wind power output values ​​of different partitions, and adding them together to obtain the wind power output time series; The historical photovoltaic output data and the capacity coefficient corresponding to the photovoltaic power source are obtained, and the photovoltaic output time series is obtained by multiplying them.

4. A method for planning a power structure considering demand-side response according to claim 3, characterized in that: After converting the original load curve into a demand response load curve according to the load demand characteristics, the method further includes: Obtaining a net load time series according to the demand response load curve and the new energy output time series; The net load time series is converted by the peak shaving and valley filling function to obtain the non-renewable energy output time series.

5. The method for planning a power structure considering demand-side response according to claim 1, characterized in that: The unit power generation costs of various power sources are calculated based on the working modes of the various power sources, including: Performing a non-sequential random production simulation according to preset constraints and the working modes of the various power sources to obtain the power generation cost; The preset constraints include power balance constraints, carbon emission constraints, system reliability constraints and buildable capacity constraints.

6. According to the method for planning a power structure considering demand-side response in claim 5, the operating modes of various power sources are obtained according to the demand response load curve and the preset objective function, including: Determine whether the current state is the initial state, and if so, generate a set of installed capacity data satisfying the constraint conditions according to the constraint conditions; If not, generating a set of optimal installed capacity data according to the objective function; The objective function is to minimize the total annual operating cost, that is: where c i is the unit power generation cost of the i-th power source, E i Its annual power generation.

7. A power structure planning device considering demand-side response, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, each step of the method for power structure planning considering demand-side response as described in any one of claims 1 to 6 is implemented.

Images