Load-following-source-moving situation awareness control method and device suitable for source-network-load-storage integration

Abstract

The invention discloses a method and a device for sensing and controlling the situation of a load moving along with a source and suitable for source-grid-load-storage integration, and the method comprises the steps: collecting historical meteorological data of a source-grid-load-storage integration energy system and corresponding fan and photovoltaic power generation data, inputting the data into a multi-time-scale aggregation model, and carrying out the prediction of the power of wind power and photovoltaic power generation; the method comprises the following steps: constructing a digital twin model for a source-network-load-storage integrated energy system, and setting equipment parameters and a production strategy of the digital twin model; taking wind power and photovoltaic power generation power data as input of the digital twinborn model, constructing a total objective function of each path of energy flow in the digital twinborn model, and predicting the distribution condition of each path of energy flow in the digital twinborn model at each moment in the future; and issuing a control instruction according to the predicted energy flow distribution condition of each path, and controlling the operation of the source-network-load-storage integrated energy system. According to the invention, the safe and stable operation of the source-grid-load-storage integrated energy system and the absorption capability of new energy power generation are improved, and the power abandoning rate of new energy is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid power balance, and in particular, relates to a load-following-source dynamic potential sensing control method and device suitable for source-grid-load-storage integration. Background Art

[0002] Traditional power grid regulation mainly adopts the mode of "source follows load". When the power load increases suddenly, once the power generation capacity on the power supply side is insufficient, there will be an imbalance between supply and demand, affecting the safe operation of the power grid. With the acceleration of the construction of new power systems, the proportion of new energy represented by wind power and photovoltaic power in the energy system structure continues to increase. Its volatility, intermittency and randomness have brought challenges to the safe and stable operation of the power grid. Therefore, the overall solution of source, grid, load and storage integration has come into being.

[0003] Source-grid-load-storage integration is a new type of power operation mode. It plans and operates power sources, power grids, loads and energy storage as a whole, which helps to improve the new energy absorption capacity and the safe and stable operation level of the power system, thereby effectively solving the problem of power system frequency and voltage fluctuations caused by the gradual increase in the proportion of new energy power generation output, improving the ability to absorb new energy power generation and improving the safe and stable operation level of regional power grids.

[0004] As the proportion of new energy power generation output gradually increases, compared with traditional thermal power, nuclear power, and hydropower units, the power generation capacity of new energy such as wind and solar power is uncontrollable. Therefore, the traditional "source follows load" mode no longer meets the demand, and the exploration of a new control mode such as "load follows source" is the general trend. Load follows source means that the load in the power system fluctuates as much as possible with the output fluctuations of wind and solar power. In the source-grid-load-storage integrated energy system, the power supply side is often dominated by new energy such as wind power and photovoltaic power generation, supplemented by large power grids or other power sources such as diesel generators. The power generation of wind power and photovoltaic power is heavily dependent on natural meteorological resources, and has the characteristics of volatility, intermittency and randomness. Therefore, there are great technical difficulties in load following source, including: it is difficult to match power demand and supply, renewable energy fluctuations and load changes are easy to impact the stability of the power grid, load tracking and energy scheduling mechanisms are more complex and have higher requirements for accuracy and real-time performance, etc., which often lead to problems such as difficulty in real-time power balance, instability of the power system, low utilization rate of new energy, and poor load operation production capacity. Summary of the invention

[0005] In response to the problems existing in the prior art, the present invention provides a load-follow-source dynamic potential sensing control method and device suitable for source-grid-load-storage integration. On the one hand, it helps to solve the power balance problem caused by power mismatch between the power supply side and the load side, and improve the safe and stable operation level of the source-grid-load-storage integrated energy system; on the other hand, it helps to improve the absorption capacity of new energy power generation, thereby reducing the new energy power abandonment rate and improving the load operation production capacity.

[0006] To achieve the above technical objectives, the present invention adopts the following technical solution: a load-following source dynamic potential sensing control method applicable to source-grid-load-storage integration, specifically comprising the following steps:

[0007] Step S1, collect historical meteorological data of the source-grid-load-storage integrated energy system and the corresponding wind turbine and photovoltaic power generation data, input them into a multi-time scale aggregation model, perform wind power and photovoltaic power generation power forecasting, and draw wind power and photovoltaic power generation power forecasting curves;

[0008] Step S2: construct a digital twin model for the source-grid-load-storage integrated energy system, and set equipment parameters and production strategies for the digital twin model;

[0009] Step S3: Taking the wind power and photovoltaic power generation power prediction curves as the input of the digital twin model, constructing the total objective function of each energy flow in the digital twin model, and optimizing and solving the optimal allocation of each energy flow in the digital twin model at each moment in the future according to the set equipment parameters and production strategies and taking minimization of the total objective function as the goal;

[0010] Step S4: issuing control instructions based on the predicted optimal distribution of each energy flow to control the operation of the source-grid-load-storage integrated energy system.

[0011] Furthermore, step S1 includes the following sub-steps:

[0012] Step S1.1, the multi-time scale aggregation model is composed of a statistical model, a machine learning model and a deep learning model in parallel;

[0013] Step S1.2, inputting historical wind turbine or photovoltaic power generation data into the statistical model respectively, and predicting the wind power or photovoltaic power generation power including the time series;

[0014] Step S1.3, inputting historical meteorological data and corresponding wind turbine or photovoltaic power generation data into a machine learning model to predict wind power or photovoltaic power generation including time series;

[0015] Step S1.4, inputting historical meteorological data and corresponding wind turbine or photovoltaic power generation data into the deep learning model to predict the wind power or photovoltaic power generation power including the time series;

[0016] Step S1.5: weight the wind power generation power predicted by the statistical model, the machine learning model and the deep learning model to obtain the wind power generation power prediction result, and draw it into a wind power generation power prediction curve; weight the photovoltaic power generation power predicted by the statistical model, the machine learning model and the deep learning model to obtain the photovoltaic power generation power prediction result, and draw it into a photovoltaic power generation power prediction curve.

[0017] Furthermore, the digital twin model in step 2 includes: a power balance module and a chemical material balance module connected in series; the power balance module includes: a green power unit, an electric energy storage unit, a direct production unit, a hydrogen production load unit and a public works power unit, the green power unit is used to store the predicted wind power and photovoltaic power generation power as the energy of the source-grid-load-storage integrated energy system through the corresponding energy flow to the electric energy storage unit and flow to the direct production unit for production, the electric energy storage unit and the energy in the direct production unit flow to the hydrogen production load unit and the public works power unit respectively through the corresponding energy flow Public engineering power unit; the chemical material balance module includes: a hydrogen storage tank, an ammonia production line, a methanol production line, an ammonia production unit and a methanol production unit. The energy in the hydrogen production load unit flows to the hydrogen storage tank through energy flow, the energy in the hydrogen storage tank and the electric energy storage unit, and the energy directly used in the production unit flows to the ammonia production line and the methanol production line through energy flow, the energy in the ammonia production line flows to the ammonia production unit through energy flow, the energy in the methanol production line flows to the methanol production unit through energy flow, and the energy in the public engineering power unit flows to the ammonia production and methanol production links through corresponding energy flows.

[0018] Furthermore, the energy flow entering the direct production unit has a higher priority than the energy flow entering the electric energy storage unit, and the energy flow entering the public works power unit has a higher priority than the energy flow entering the hydrogen production load unit. When the energy of the battery energy unit is insufficient to supply the hydrogen production load unit through the corresponding energy flow, it is supplied through the public works power unit.

[0019] Furthermore, it also includes an off-grid power unit, an abandoned power unit and an additional hydrogen supply unit. When the energy provided by the green power unit is insufficient, the off-grid power unit is used to supplement energy; when the energy provided by the green power unit exceeds the energy storage capacity of the electric energy storage unit, the excess energy is discarded through the abandoned power unit; when the energy required for the ammonia production line and the methanol production line is insufficient, hydrogen is supplied through the additional hydrogen supply unit.

[0020] Furthermore, the equipment parameters of the digital twin model set in step S2 include: setting the energy storage capacity, charging and discharging power, charging and discharging loss, discharge depth and initial energy storage charge of the electric energy storage unit, setting the power, hydrogen production capacity and load adjustment range of the hydrogen production load unit, setting the number of hydrogen storage tanks, the total hydrogen storage capacity, the hydrogen storage rate and the initial hydrogen storage capacity, setting the number of ammonia synthesis towers in the ammonia production line and the ammonia production load adjustment range, setting the number of methanol synthesis units in the methanol production line and the methanol production load adjustment range, and setting the inflow and outflow speeds, conversion ratios and losses of each energy flow.

[0021] Furthermore, the production strategies of the digital twin model set in step S2 include: power supply strategy, hydrogen supply strategy, hydrogen production cluster scheduling strategy and chemical production decision strategy.

[0022] Furthermore, the construction process of the total objective function F of each energy flow in the digital twin model in step S3 is:

[0023]

[0024] Among them, n represents the total number of energy flows in the digital twin model, i represents the index of n, flow(i,t) represents the energy flow power corresponding to the i-th energy flow at time t, and cost(i) represents the weight coefficient of the i-th energy flow.

[0025] Furthermore, if the operating time of the integrated energy system of control source, grid, load and storage in step S4 exceeds 24 hours, it is necessary to use the actual meteorological data, the generated wind power and photovoltaic power generation power, and the optimal distribution of each energy flow during the operation of the integrated energy system of control source, grid, load and storage to correct the deviation of the predicted wind power, photovoltaic power generation power curves and energy flow distribution.

[0026] Furthermore, the present invention also provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the load-following source dynamic potential sensing control method applicable to source-grid-load-storage integration is implemented.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] (1) The load-following source dynamic potential sensing control method applicable to the source-grid-load-storage integration of the present invention uses a multi-time scale aggregation model to perceive the power forecast of wind power and photovoltaic power generation. On the one hand, the multi-time scale can be combined with the short-term power forecast to improve the accuracy of the medium-term power forecast while avoiding the energy scheduling from falling into the local optimum. On the other hand, the aggregation model can avoid the problem of insufficient accuracy of a single model in dealing with different power forecasting scenarios, and enhance the ability to deal with unforeseen events such as storms and cloud cover.

[0029] (2) The present invention is applicable to the load-following source dynamic potential sensing control method of the integrated source, grid, load and storage energy system to construct a digital twin model for the integrated source, grid, load and storage energy system to simulate the operation of the entire integrated source, grid, load and storage energy system, and by constructing the total objective function of each energy flow in the digital twin model, the optimal allocation of each energy flow in the digital twin model at each moment in the future is solved by minimizing the total objective function to guide the operation of the integrated source, grid, load and storage energy system. Through optimization and solution, it can optimize the matching of loads with different priorities and power sources under multi-dimensional constraints, maximize the benefits of the integrated source, grid, load and storage energy system, and minimize the operating costs, thereby improving the economy of the integrated source, grid, load and storage energy system during its operation cycle.

[0030] (3) The present invention is applicable to the load-following source dynamic potential sensing control method of the integrated source-grid-load-storage system. On the one hand, it helps to solve the power balance problem caused by the power mismatch between the power supply side and the load side, and improve the safe and stable operation level of the integrated source-grid-load-storage energy system. On the other hand, it helps to improve the absorption capacity of new energy power generation, thereby reducing the new energy power abandonment rate and improving the load operation production capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a flow chart of the load-following source dynamic potential sensing control method applicable to the source-grid-load-storage integration of the present invention;

[0032] Figure 2 It is a topological diagram of the source-grid-load-storage integrated energy system in the present invention;

[0033] Figure 3 Schematic diagram of the digital twin model constructed for the present invention;

[0034] Figure 4 It is the electric power flow diagram of the present invention;

[0035] Figure 5 It is the energy flow diagram of chemical material transmission in the present invention;

[0036] Figure 6 is the curve diagram of the simulation of the load following the source, where: Figure 6 (a1) is the load-following source dynamic simulation curve of Example ① under the traditional control strategy. Figure 6 (a2) is the load-following source dynamic simulation curve of Example ② under the traditional control strategy. Figure 6 (b1) in the example ① is the load-following source dynamic simulation curve using only power prediction. Figure 6 (b2) in the example ② is the load-following source dynamic simulation curve using only power prediction. Figure 6 (c1) in the figure is the load-following source dynamic simulation curve of Example ① using the method of the present invention. Figure 6 (c2) in the figure is the load-following source dynamic simulation curve of example ② using the method of the present invention. DETAILED DESCRIPTION

[0037] The technical solution of the present invention is further explained below in conjunction with the accompanying drawings.

[0038] like Figure 1 The present invention is a flow chart of a load-following source dynamic potential sensing control method applicable to source-grid-load-storage integration, and the load-following source dynamic potential sensing control method specifically comprises the following steps:

[0039] Step S1, collect historical meteorological data of the source-grid-load-storage integrated energy system and the corresponding wind turbine and photovoltaic power generation data, input them into the multi-time scale aggregation model, predict wind power and photovoltaic power generation, and draw wind power and photovoltaic power generation prediction curves. On the one hand, the multi-time scale can be combined with short-term power prediction to improve the accuracy of medium-term power prediction while avoiding energy scheduling from falling into local optimality. On the other hand, the aggregation model can avoid the problem of insufficient accuracy of a single model in dealing with different power prediction scenarios, and enhance the ability to deal with unforeseen events such as storms and cloud cover, so as to accurately perceive the situation and trend of wind power and photovoltaic power generation in a period of time in the future; specifically, it includes the following sub-steps:

[0040] Step S1.1, the multi-time scale aggregation model in the present invention is composed of a statistical model, a machine learning model and a deep learning model in parallel, wherein the statistical model can adopt SARIMA or Prophet, the machine learning model can adopt RandomForest or Xgboost, and the deep learning model can adopt RNN or LSTM;

[0041] Step S1.2, inputting historical wind turbine or photovoltaic power generation data into the statistical model respectively, summarizing its regular characteristics according to statistics, and predicting the wind power or photovoltaic power generation power including the time series;

[0042] Step S1.3, inputting historical meteorological data and corresponding wind turbine or photovoltaic power generation data into the machine learning model, predicting the wind power or photovoltaic power generation including time series, converting the time series problem into supervised learning, which can solve most of the complex time series prediction problems and avoid manual intervention;

[0043] Step S1.4, inputting historical meteorological data and corresponding wind turbine or photovoltaic power generation data into the deep learning model to predict the wind, solar or photovoltaic power generation power including time series. Compared with traditional neural networks, it can effectively capture important short-term and long-term dependencies in time series and solve the problems of gradient disappearance and gradient explosion;

[0044] Step S1.5: Weight the wind power generation power predicted by the statistical model, machine learning model and deep learning model to obtain the wind power generation power prediction result, and draw it into a wind power generation power prediction curve; weight the photovoltaic power generation power predicted by the statistical model, machine learning model and deep learning model to obtain the photovoltaic power generation power prediction result, and draw it into a photovoltaic power generation power prediction curve. The aggregation model will train the model according to the prediction results of various models and the actual power generation, and continuously and adaptively adjust the weights of the prediction results of different models to continuously optimize the accuracy of the final prediction results.

[0045] Step S2: Figure 2 It is a topological diagram of the source-grid-load-storage integrated energy system in the present invention. According to the topological diagram, a digital twin model of the source-grid-load-storage integrated energy system is constructed, and the equipment parameters and production strategies of the digital twin model are set.

[0046] like Figure 3 The digital twin model in the present invention includes: a power balance module and a chemical material balance module connected in series, wherein Figure 3 The red frame in the figure indicates the power balance module. Figure 3 The blue frame part in the figure represents the chemical material balance module; the power balance module includes: green power unit, electric energy storage unit, direct production unit, hydrogen production load unit and public engineering power unit. The green power unit is used to store the predicted wind power and photovoltaic power generation power as the energy of the source-grid-load-storage integrated energy system through the corresponding energy flow to the electric energy storage unit and flow to the direct production unit for production. The energy in the electric energy storage unit and the direct production unit flows to the hydrogen production load unit and the public engineering power unit respectively through the corresponding energy flow; the chemical material balance module includes: hydrogen storage tank, ammonia production line, methanol production line, ammonia production unit and methanol production unit. The energy in the hydrogen production load unit flows to the hydrogen storage tank through the energy flow, the energy in the hydrogen storage tank and the electric energy storage unit and the direct production unit flows to the ammonia production line and the methanol production line through the energy flow, the energy in the ammonia production line flows to the ammonia production unit through the energy flow, the energy in the methanol production line flows to the methanol production unit through the energy flow, and the energy in the public engineering power unit flows to the ammonia production and methanol production links through the corresponding energy flow.

[0047] The digital twin model of the present invention also includes an off-grid power unit, an abandoned power unit and an additional hydrogen supply unit. When the energy provided by the green power unit is insufficient, the off-grid power unit is used to supplement energy; when the energy provided by the green power unit exceeds the energy storage capacity of the electric energy storage unit, the excess energy is discarded through the abandoned power unit; when the energy required for the ammonia production line and the methanol production line is insufficient, hydrogen is supplied by the additional hydrogen supply unit. When there is no additional hydrogen supply unit to supply hydrogen, production should be stopped.

[0048] The energy flow entering the direct production unit has a higher priority than the energy flow entering the electric energy storage unit. The energy flow entering the public engineering power unit has a higher priority than the energy flow entering the hydrogen production load unit. When the energy of the battery energy unit is insufficient to supply the hydrogen production load unit through the corresponding energy flow, it is supplied through the public engineering power unit.

[0049] in addition, Figure 3 middle It is the input end, that is, it provides energy to the outside. When the value of this component has been set, energy will be provided according to the corresponding value, and this energy must have a destination in the energy system; when the value of this component has not been set, it will automatically provide appropriate energy according to the balance of the energy system and its weight in combination with the production strategy. It is the output end, that is, it receives energy. When the value of this component has been set, it will absorb energy according to the corresponding value. If it cannot absorb enough energy, the production strategy will not be met. When the value of this component has not been set, it will automatically absorb appropriate energy according to the balance of the energy system and its weight in combination with the production strategy. For energy storage unit, For middleware, As a busbar, it can ensure that the incoming energy flow and the outgoing energy flow are always consistent.

[0050] The equipment parameters of the digital twin model set in the present invention include: setting the energy storage capacity, charging and discharging power, charging and discharging loss, discharge depth and initial energy storage charge of the electric energy storage unit, setting the power, hydrogen production capacity and load adjustment range of the hydrogen production load unit, setting the number of hydrogen storage tanks, the total hydrogen storage capacity, the hydrogen storage rate and the initial hydrogen storage capacity, setting the number of ammonia synthesis towers in the ammonia production line and the ammonia production load adjustment range, setting the number of methanol synthesis units in the methanol production line and the methanol production load adjustment range, and setting the inflow and outflow speeds, conversion ratios and losses of each energy flow.

[0051] The production strategies of the digital twin model set up in the present invention include: power supply strategy, hydrogen supply strategy, hydrogen production cluster scheduling strategy and chemical production decision-making strategy.

[0052] Step S3, using the wind power and photovoltaic power generation power prediction curves as the input of the digital twin model, constructing the total objective function of each energy flow in the digital twin model, and according to the set equipment parameters and production strategies, taking minimization of the total objective function as the goal, using the mixed integer linear programming MILP algorithm and solver to calculate and solve the optimal allocation of each energy flow in the digital twin model at each moment in the future. Through the optimization solution, the matching of loads with different priorities and power sources can be optimized under multi-dimensional constraints, and the benefits of the integrated energy system of source, grid, load and storage can be maximized and the operating costs can be minimized, thereby improving the economy of the integrated energy system of source, grid, load and storage during the operation cycle.

[0053] The construction process of the total objective function F of each energy flow in the digital twin model of the present invention is:

[0054]

[0055] Among them, n represents the total number of energy flows in the digital twin model, i represents the index of n, flow(i,t) represents the energy flow power corresponding to the i-th energy flow at time t, and cost(i) represents the weight coefficient of the i-th energy flow. The weight coefficient of the energy flow determines the priority of power supply and load and the priority of energy allocation during the operation of the energy system.

[0056] Step S4, issue control instructions according to the predicted energy flow distribution of each route to control the operation of the source-grid-load-storage integrated energy system, including the wind turbine control system, photovoltaic control system, energy storage charging and discharging control system, hydrogen production and storage control system and ammonia chemical control system, and ultimately achieve efficient operation, safety and stability of the entire source-grid-load-storage integrated energy system.

[0057] like Figure 4 and Figure 5 As shown in the figure, through the simulation calculation of the digital twin model, the real-time values ​​of the power transmission and material transmission between the various components in the source-grid-load-storage integrated energy system can be clearly displayed, and the operation of the entire source-grid-load-storage integrated energy system can be intuitively reflected. At the same time, various key indicators can be extracted from the simulation results, such as the green power abandonment rate, the power grid purchase rate, and the annual total output, to quantify the overall operation performance of the source-grid-load-storage integrated energy system.

[0058] In order to analyze and compare the improvement effect of the situation awareness control method proposed in the present invention, according to two specific examples, example ① is a wind-hydrogen-alcohol storage project with 2.1GW wind power, 300MW / 300MWh energy storage, 280 hydrogen electrolyzers, and an alcohol production scale of 800,000 tons / year, and example ② is a wind-solar-hydrogen-ammonia storage project with 1GW wind power, 300MW photovoltaic, 200MW / 200MWh energy storage, 132 hydrogen electrolyzers, and a synthetic ammonia scale of 450,000 tons / year. (a) The traditional control strategy, (b) the traditional control strategy combined with power prediction, and (c) the method of the present invention are simulated and calculated, and the results are shown as follows. Figure 6 The simulation results under different control strategies are shown in Table 1. Figure 6 It can be seen intuitively from the simulation curves of the two groups of examples that after adding the power prediction function, Figure 6 As shown in (b1) and (b2) in Fig. Figure 6 The traditional control strategies of (a1) and (a2) have been significantly improved. Figure 6Directly using the method of the present invention in (c1) and (c2) can further improve the tracking effect of the load-following source. From the data analysis in Table 1, it can be seen that the various indicators of method (c) are the best, method (b) is second, and method (a) performs the worst. Compared with control strategy (a), the new energy power abandonment rate of method (c) is reduced by about 40%, the annual load shutdown times are reduced by about 20%, and the maximum power deviation is reduced by about 60%. The above results show that the load-following source dynamic potential sensing control method proposed in the present invention, on the one hand, helps to solve the power balance problem caused by the power mismatch between the power supply side and the load side, and improves the safe and stable operation level of the source-grid-load-storage integrated energy system; on the other hand, it helps to improve the absorption capacity of new energy power generation, thereby reducing the new energy power abandonment rate and improving the load operation production capacity.

[0059] Table 1 Simulation results under different control strategies

[0060]

[0061] In one technical solution of the present invention, if the operating time of the integrated energy system of control source, grid, load and storage in step S4 exceeds 24 hours, it is necessary to use the actual meteorological data, the generated wind power and photovoltaic power generation power and the optimal distribution of each energy flow during the operation of the integrated energy system of control source, grid, load and storage to correct the deviation of the predicted wind power, photovoltaic power generation power curve and the energy flow distribution.

[0062] In a technical solution of the present invention, an electronic device is also provided, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the load-following source dynamic potential sensing control method applicable to source-grid-load-storage integration is implemented.

[0063] In the embodiments disclosed in the present application, the computer storage medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. The computer storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the above. More specific examples of computer storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above.

[0064] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0065] The above are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should be regarded as the protection scope of the present invention.

Claims

1. A load-following source dynamic potential sensing control method suitable for source-grid-load-storage integration, characterized in that: The specific steps include: Step S1, collect historical meteorological data of the source-grid-load-storage integrated energy system and the corresponding wind turbine and photovoltaic power generation data, input them into a multi-time scale aggregation model, perform wind power and photovoltaic power generation power forecasting, and draw wind power and photovoltaic power generation power forecasting curves; Step S2: construct a digital twin model for the source-grid-load-storage integrated energy system, and set equipment parameters and production strategies for the digital twin model; Step S3: Taking the wind power and photovoltaic power generation power prediction curves as the input of the digital twin model, constructing the total objective function of each energy flow in the digital twin model, and optimizing and solving the optimal allocation of each energy flow in the digital twin model at each moment in the future according to the set equipment parameters and production strategies and taking minimization of the total objective function as the goal; Step S4: issuing control instructions according to the predicted optimal distribution of each energy flow to control the operation of the source-grid-load-storage integrated energy system.

2. The load-following source dynamic potential sensing control method applicable to source-grid-load-storage integration according to claim 1 is characterized in that: Step S1 includes the following sub-steps: Step S1.1, the multi-time scale aggregation model is composed of a statistical model, a machine learning model and a deep learning model in parallel; Step S1.2, inputting historical wind turbine or photovoltaic power generation data into the statistical model respectively, and predicting the wind power or photovoltaic power generation power including the time series; Step S1.3, inputting historical meteorological data and corresponding wind turbine or photovoltaic power generation data into a machine learning model to predict wind power or photovoltaic power generation including time series; Step S1.4, inputting historical meteorological data and corresponding wind turbine or photovoltaic power generation data into the deep learning model to predict the wind power or photovoltaic power generation power including the time series; Step S1.5: weight the wind power generation power predicted by the statistical model, the machine learning model and the deep learning model to obtain the wind power generation power prediction result, and draw it into a wind power generation power prediction curve; weight the photovoltaic power generation power predicted by the statistical model, the machine learning model and the deep learning model to obtain the photovoltaic power generation power prediction result, and draw it into a photovoltaic power generation power prediction curve.

3. A load-following source dynamic potential sensing control method applicable to source-grid-load-storage integration according to claim 2, characterized in that: The digital twin model in step 2 includes: a power balance module and a chemical material balance module connected in series; the power balance module includes: a green power unit, an electric energy storage unit, a direct production unit, a hydrogen production load unit and a public works power unit, the green power unit is used to store the predicted wind power and photovoltaic power generation power as the energy of the source-grid-load-storage integrated energy system through the corresponding energy flow to the electric energy storage unit and flow to the direct production unit for production, the electric energy storage unit and the energy in the direct production unit flow to the hydrogen production load unit and the public works respectively through the corresponding energy flow The chemical material balance module includes: a hydrogen storage tank, an ammonia production line, a methanol production line, an ammonia production unit and a methanol production unit. The energy in the hydrogen production load unit flows to the hydrogen storage tank through the energy flow. The energy in the hydrogen storage tank, the electric energy storage unit and the energy directly used in the production unit flow to the ammonia production line and the methanol production line through the energy flow. The energy in the ammonia production line flows to the ammonia production unit through the energy flow. The energy in the methanol production line flows to the methanol production unit through the energy flow. The energy in the public engineering power unit flows to the ammonia production and methanol production links through the corresponding energy flow.

4. The load-following source dynamic potential sensing control method applicable to source-grid-load-storage integration according to claim 3 is characterized in that: The energy flow entering the direct production unit has a higher priority than the energy flow entering the electric energy storage unit. The energy flow entering the public engineering power unit has a higher priority than the energy flow entering the hydrogen production load unit. When the energy of the battery energy unit is insufficient to supply the hydrogen production load unit through the corresponding energy flow, it is supplied through the public engineering power unit.

5. The load-following source dynamic potential sensing control method applicable to source-grid-load-storage integration according to claim 4 is characterized in that: It also includes off-grid power units, abandoned power units and additional hydrogen supply units. When the energy provided by the green power unit is insufficient, the off-grid power unit is used to supplement energy; when the energy provided by the green power unit exceeds the energy storage capacity of the electric energy storage unit, the excess energy is discarded through the abandoned power unit; when the energy required for the ammonia production line and the methanol production line is insufficient, hydrogen is supplied through the additional hydrogen supply unit.

6. The load-following source dynamic potential sensing control method applicable to source-grid-load-storage integration according to claim 5 is characterized in that: The equipment parameters of the digital twin model set in step S2 include: setting the energy storage capacity, charging and discharging power, charging and discharging loss, discharge depth and initial energy storage charge of the electric energy storage unit, setting the power, hydrogen production capacity and load adjustment range of the hydrogen production load unit, setting the number of hydrogen storage tanks, the total hydrogen storage capacity, the hydrogen storage rate and the initial hydrogen storage capacity, setting the number of ammonia synthesis towers in the ammonia production line and the ammonia production load adjustment range, setting the number of methanol synthesis units in the methanol production line and the methanol production load adjustment range, and setting the inflow and outflow speeds, conversion ratios and losses of each energy flow.

7. The load-following source dynamic potential sensing control method applicable to source-grid-load-storage integration according to claim 6 is characterized in that: The production strategies of the digital twin model set in step S2 include: power supply strategy, hydrogen supply strategy, hydrogen production cluster scheduling strategy and chemical production decision strategy.

8. The load-following source dynamic potential sensing control method applicable to source-grid-load-storage integration according to claim 7 is characterized in that: The construction process of the total objective function F of each energy flow in the digital twin model in step S3 is: Among them, n represents the total number of energy flows in the digital twin model, i represents the index of n, flow(i,t) represents the energy flow power corresponding to the i-th energy flow at time t, and cost(i) represents the weight coefficient of the i-th energy flow.

9. The load-following source dynamic potential sensing control method applicable to source-grid-load-storage integration according to claim 1 is characterized in that: If the operation time of the integrated energy system of control source, grid, load and storage in step S4 exceeds 24 hours, it is necessary to use the actual meteorological data, the generated wind power and photovoltaic power generation power and the optimal distribution of each energy flow during the operation of the integrated energy system of control source, grid, load and storage to correct the deviation of the predicted wind power, photovoltaic power generation power curve and the energy flow distribution.

10. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the load-following source dynamic potential sensing control method applicable to source-grid-load-storage integration as described in any one of claims 1 to 9 is implemented.

Images