Energy system optimization scheduling method considering flexible load

Abstract

The invention discloses an energy system optimization scheduling method considering flexible loads, which comprises the following steps: S1, by using new technologies such as intelligent home appliance networking, electric vehicle ordered charging and household energy storage in a transformer area, overall management of transformer area flexible loads such as electric vehicles, household energy storage and intelligent Internet of Things home appliances is carried out, and load adjustment characteristics are analyzed, so that the intelligent home appliance networking of the transformer area is realized; establishing a load aggregation model; s2, then formulating a transformer area side flexible load collaborative optimization scheduling strategy; s3, considering a scene that multiple transformer areas participate in power grid dispatching, establishing a regulation and control framework, and proposing a dispatching strategy under multiple time scales; and S4, mining the flexible load resource regulation and control capability of the transformer area, and achieving the goals of optimal power utilization of the transformer area, stabilization of new energy fluctuation, safe, stable and economical operation of a power grid and the like. According to the energy system optimization scheduling method considering the flexible load, typical flexible load adjustment characteristics are analyzed, a typical flexible load aggregation model is established, and the problem that the flexible load adjustment characteristics are difficult to analyze due to limitation of influence factors is solved.

Description

Technical Field

[0001] The present invention relates to the technical field related to energy system optimization scheduling, and specifically to an energy system optimization scheduling method considering flexible loads. Background Art

[0002] During peak hours of electricity consumption, the excessive load on the power grid affects the electricity consumption of users and their production and life. Therefore, it is necessary to optimize the scheduling of the energy system. However, the existing energy system optimization scheduling method considering flexible loads still has certain defects when used, such as;

[0003] In the process of use, there are many types of energy system optimization scheduling methods and the operation strategy is not unique, so the overall processing is relatively complicated, which makes the optimization model calculation time long and difficult to apply in real time, and cannot effectively improve the flexibility of system operation. In addition, high-precision scheduling requires a large amount of real-time data and forecast data, which puts high demands on data collection and processing, and increases implementation costs;

[0004] In order to overcome the above defects, prior art 1 (Chinese patent with application number 201910192498.9 and application date 2019-03-14) is a method for optimizing the dispatching of an integrated energy system taking into account multiple flexible loads. Combined with the power consumption and dispatching characteristics of various flexible loads in the integrated energy system, a mathematical model of the integrated energy system including a load model that can be translated, transferred, and reduced, energy storage, and a gas turbine power generation system is established according to an incentive compensation mechanism; based on an integrated energy system in which both electric and thermal loads can participate in the interaction, a combined day-ahead optimization dispatching model of supply and demand of the integrated energy system is established with the goal of minimizing the total operating cost, and the model is imported into an automated computer system. The automated computer system solves the output of each device in the integrated energy system, controls each device according to the output, and obtains an optimized operation result;

[0005] There is also prior art 2 (Chinese patent with application number 201911107054.7 and application date 2019-11-13), a method for optimizing and dispatching a regional integrated energy system considering flexible heat load, which first obtains basic data of the regional integrated energy system to be optimized, and builds a flexible heat load model considering the user's work and rest rules based on the basic data obtained, and then builds a regional integrated energy system day-ahead optimization dispatching model with minimum energy cost as the objective function and electric power balance constraint, thermal power balance constraint, energy storage device storage capacity constraint, etc. as constraints based on the above model, and then solves the optimization dispatching model to obtain the input and output of various equipment in the optimized regional integrated energy system and the user's indoor temperature change curve, and finally dispatches various equipment according to the above optimization scheme;

[0006] And prior art 3 (Chinese patent with application number 201811322703.0 and application date 2018-11-07) is an optimization scheduling method for an integrated energy system considering carbon trading mechanism and flexible load. Based on the energy center model, the coupling links of the integrated energy system are modeled, the input energy is electricity and natural gas, and the output is electricity, heat and natural gas to supply the load demand; the carbon trading mechanism is introduced into the scheduling model of the integrated energy system, the carbon emissions of the entire energy supply link are comprehensively considered, and an over-emission penalty mechanism is established; the flexible load is applied to the low-carbon economic operation of the integrated energy system; on the above basis, the power system network constraints, the natural gas system network constraints and the internal constraints of the energy center are introduced, and a low-carbon economic scheduling model of the integrated energy system is established with carbon trading costs, flexible load scheduling costs and system energy consumption costs as targets.

[0007] Although existing technologies can set up models for optimization, during the operation of the system, the overall energy is limited by the type of flexible load, the basic operating conditions and related parameters inside the equipment, and the user's electricity consumption expectations, which makes it difficult to analyze the flexible load regulation characteristics. In addition, since the flexible load access locations are scattered and most of them are small and medium-sized loads with small individual capacities, they cannot be effectively aggregated to form economies of scale, making it inconvenient to establish a flexible load aggregation model, which lays a poor foundation for flexible load participation in scheduling.

[0008] In response to the above problems, it is urgently necessary to make innovative designs based on the original energy system optimization scheduling method considering flexible loads. Therefore, we proposed an energy system optimization scheduling method considering flexible loads that can solve the above problems well. Summary of the invention

[0009] The purpose of the present invention is to provide an energy system optimization scheduling method taking flexible loads into consideration, so as to solve the problem raised by the above-mentioned background technology that although a model can be set up for optimization in the current market, during the operation of the system, the overall energy is restricted by the type of flexible load, the basic operating conditions and related parameters inside the equipment, and the user's electricity consumption expectations, which makes it difficult to analyze the flexible load regulation characteristics. In addition, since the flexible loads are accessed at dispersed locations and are mostly small and medium-sized loads with small individual capacities, they cannot be effectively aggregated to form economies of scale, making it inconvenient to establish a flexible load aggregation model, which makes it difficult to lay a good foundation for the flexible loads to participate in scheduling.

[0010] To achieve the above object, the present invention provides the following technical solution: an energy system optimization scheduling method considering flexible loads, comprising the following steps:

[0011] S1: Utilize new technologies such as smart home appliance networking, orderly charging of electric vehicles, and household energy storage to coordinate the management of flexible loads in the substation, such as electric vehicles, household energy storage, and smart IoT appliances. Establish a load aggregation model by analyzing the load regulation characteristics.

[0012] S2: Then formulate the coordinated optimization dispatch strategy of flexible load on the substation side;

[0013] S3: Considering the scenario where multiple substations participate in grid dispatching, a control framework is established and dispatching strategies under multiple time scales are proposed;

[0014] S4: Starting from the new background and new situation, using emerging technologies, we will tap into the flexible load resource regulation capabilities of the substation, achieve the goals of optimal power consumption in the substation, smooth out fluctuations in new energy, and ensure safe, stable and economical operation of the power grid. We will increase the new energy absorption rate, protect the vital interests of users, and provide new ideas for optimal operation management and energy efficiency improvement of future distribution network substations.

[0015] Preferably, typical flexible loads of users in the substation include temperature control loads, electric vehicles and energy storage equipment. Since flexible loads have good adjustment characteristics, typical types of flexible loads are considered to participate in the substation optimization scheduling.

[0016] Preferably, the temperature load adopts temperature control to achieve more refined regulation. The temperature control of the temperature-controlled load is mainly based on the changes in electricity prices and outdoor temperature. The load is transferred from high electricity price periods to low electricity price periods to reduce user costs, or the temperature setting value is adjusted according to the system operation conditions to provide auxiliary services for the system and ensure the stable operation of the power system.

[0017] Preferably, the load aggregator controls the temperature-controlled load by means of temperature control, taking into account the benefits generated by the control behavior and giving users certain subsidies. When the load aggregator's interests are maximized, the optimization goal is:

[0018]

[0019] Where G represents the interest of the load aggregator.

[0020] Preferably, users are often concerned about their own comfort. The comfort of temperature control load is mainly related to temperature. Therefore, the user level takes the minimum temperature change as the optimization goal, which is expressed as follows:

[0021]

[0022] Wherein, Tk set represents the temperature setting value; T0 set represents the temperature expected value.

[0023] Preferably, the aggregate power constraint condition is:

[0024]

[0025] Where P dr,t,min and P dr,t,max They respectively represent the minimum and maximum values ​​of the demand-side response. When the temperature changes by more than 3°C, the human body will obviously feel uncomfortable, so the temperature change needs to be constrained within 3°C. Finally, the intelligent optimization algorithm is used to obtain the temperature control load optimization scheduling strategy.

[0026] Preferably, the time-optimized charging and discharging amount is used as the connecting variable for the upper-layer time scheduling optimization and the lower-layer space scheduling optimization. When performing real-time space allocation scheduling, the number of charging vehicles is allocated according to the charging and discharging amount in each time period of the daily plan to achieve a balance between supply and demand of electricity.

[0027] Preferably, under the premise of ensuring that the dispatchable capacity of user-side energy storage and the energy storage capacity of aggregation side meet the peak-shaving demand of the system, the PC / EC configuration of the energy storage capacity on the aggregation side is studied with the goal of maximizing the annual profit of the aggregator and minimizing the peak-to-valley difference of the electricity consumption curve. The profit of the aggregator is the profit F1 when participating in peak-shaving, and the expenditure of the aggregator is divided into three parts.

[0028] Preferably, the first part is the cost of aggregating user-side energy storage F2, the second part is the cost of configuring and maintaining the aggregate-side energy storage F3, and the third part is the charging cost of the aggregate-side energy storage F4. Therefore, with the goal of maximizing the annual revenue of the aggregator and minimizing the peak-to-valley difference of the electricity consumption curve, the objective function for configuring the aggregate-side energy storage capacity is:

[0029]

[0030]

[0031] Considering the peak-shaving power and capacity constraints, the power balance constraints of the substation, and the energy storage operation constraints, the optimization algorithm is used to obtain the optimal scheduling strategy for energy storage equipment.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows: the energy system optimization scheduling method considering flexible loads analyzes the typical flexible load regulation characteristics, establishes a typical flexible load aggregation model, and solves the problem that the flexible load regulation characteristics are difficult to analyze due to the limitation of influencing factors. The specific contents are as follows:

[0033] (1) Analyze the regulation characteristics of typical flexible loads, fully consider the various factors affecting the operation status and usage time of typical flexible loads, mathematically process these factors and integrate them into the load model, and finally aggregate them to establish a typical flexible load aggregation model. This solves the problem that the regulation characteristics of flexible loads are difficult to analyze due to many influencing factors such as the type of flexible loads, the basic operating conditions and related parameters of the equipment used, and the user's electricity consumption expectations;

[0034] (2) Use load aggregators to achieve unified management of flexible loads in the substation. For typical flexible loads, the optimal scheduling strategy for typical flexible loads of users in the substation is proposed with the goal of maximizing the interests of the aggregator and optimizing the comfort of users, taking into account the constraints of the load itself, the constraints of user power consumption, and the balance constraints of supply and use.

[0035] (3) On the basis of unified management, considering the coupling and randomness of the use time of various typical flexible loads, as well as the changes in the operating status of the loads themselves, an optimal strategy for joint dispatch of multiple flexible loads of users in the substation area is proposed;

[0036] (4) According to the different loads within users, the potential characteristics and dispatching features of the users that can participate in the grid dispatching are analyzed, and a dispatching framework of the centralized aggregator model for flexible loads to participate in the dispatching is proposed. A multi-time scale dispatching strategy for the access of new energy to the distribution network of multiple distribution areas is formulated. The game behavior of multiple distribution areas is considered, and the uncertainty of source, load and storage is quantified. Under the constraints of grid power balance, node voltage constraints, energy storage constraints, flexible load constraints, active and reactive output, etc., the maximum consumption of new energy and the optimal economic benefits of the grid are taken as the optimization goals, and the day-ahead and real-time dispatching strategies are obtained;

[0037] (5) Active distribution network optimization scheduling based on flexible loads is realized on multiple time scales, and the use of flexible loads is implemented to smooth the fluctuations of renewable energy, solving the problem of real-time scheduling difficulties caused by the uncertainty of sources, loads and storage. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic diagram of the structure of the ETP model diagram of the present invention;

[0039] Figure 2 It is the basic flow chart of the NSGA-II algorithm of the present invention;

[0040] Figure 3 This is a schematic diagram of the classification of game theory of the present invention;

[0041] Figure 4 It is a schematic diagram of the relationship between various topics of the present invention;

[0042] Figure 5 This is a user-level technical framework diagram of the present invention;

[0043] Figure 6 This is a technical framework diagram of the station area-user level of the present invention;

[0044] Figure 7 This is the technical roadmap of the distribution network-area level of the present invention;

[0045] Figure 8 It is a schematic diagram of a general parameter serialization control method of the present invention;

[0046] Fig. 9 A schematic diagram of energy flow and information flow between the three entities of the present invention;

[0047] Fig.10 Solve the flow chart for the present invention;

[0048] Fig.11 This is the intraday rolling optimization strategy diagram of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] Embodiment 1:

[0051] Typical flexible loads of users in the substation include temperature control loads, electric vehicles and energy storage equipment. Since flexible loads have good adjustment characteristics, typical types of flexible loads are considered to participate in the substation optimization scheduling, such as Figure 4-Figure 7 ;

[0052] Temperature control load:

[0053] Temperature load adopts temperature control to achieve more refined regulation. The temperature control of temperature-controlled load is mainly based on the changes in electricity prices and outdoor temperature. The load is transferred from high electricity price periods to low electricity price periods to reduce user costs, or the temperature setting value is adjusted according to the system operation status to provide auxiliary services for the system and ensure the stable operation of the power system;

[0054] The load aggregator controls the temperature-controlled load by means of temperature control. Considering the benefits of the control behavior and giving users certain subsidies, the optimization goal is to maximize the interests of the load aggregator:

[0055]

[0056] Where G represents the interest of the load aggregator;

[0057] Users are often concerned about their own comfort. The comfort of temperature control load is closely related to temperature. Therefore, the user level takes the minimum temperature change as the optimization goal, which is expressed as follows:

[0058]

[0059] In the formula, Tk set represents the temperature setting value; T0 set represents the temperature expected value;

[0060] The aggregate power constraints are:

[0061]

[0062] Where P dr,t,min and P dr,t,max They represent the minimum and maximum values ​​of the demand-side response respectively. When the temperature changes by more than 3°C, the human body will obviously feel uncomfortable, so the temperature change needs to be constrained within 3°C. Finally, the intelligent optimization algorithm is used to obtain the temperature control load optimization scheduling strategy.

[0063] Electric Vehicles:

[0064] The implementation of V2G technology is related to the type, quantity and charging method of electric vehicles. The implementation of V2G can be divided into the following types:

[0065] Centralized V2G refers to the centralized management of all electric vehicles in a certain area. By formulating specific dispatching strategies and using intelligent algorithms to calculate the optimal charging and discharging strategy for each electric vehicle, the energy stored in electric vehicles can be automatically dispatched in a unified manner when the grid load is insufficient, achieving goals such as smoothing the grid load curve.

[0066] Autonomous V2G: A single electric vehicle is generally automatically connected to the power grid through an on-board smart charger. Charging and discharging are not restricted by time and place. The electric vehicle can be guided to charge and discharge based on information commands issued by the power grid, or signals such as the electrical characteristics of the power grid output interface, combined with its own power status and related electricity price incentives. Autonomous V2G requires more sensitive communication equipment and on-board smart chargers, which will increase the cost of electric vehicles. Since electric vehicles are not subject to unified scheduling, the charging and discharging methods are flexible and convenient. Therefore, the realization of V2G is very random, and reverse peaks may occur, which may not necessarily achieve overall V2G.

[0067] Based on microgrid V2G, the service object of microgrid V2G is microgrid, which mainly includes distributed power generation such as wind power and solar energy. Therefore, due to regional restrictions, electric vehicles are dispatched by microgrids. When the power of microgrid is insufficient, electric vehicles begin to discharge for related loads. When the power of microgrid is in excess, electric vehicles are charged. Microgrid V2G can also connect and communicate with external large power grids to provide power to users;

[0068] Battery-replacement V2G: A large number of spare batteries are stored inside the battery swap station. Some batteries are kept fully charged to meet the needs of electric vehicle users for quick battery replacement when traveling; the other batteries can participate in grid interaction, enabling the battery pack to interact with the grid as an energy storage unit to achieve the purpose of peak load reduction and valley filling.

[0069] The time-space dual-dimensional scheduling model for orderly charging and discharging of electric vehicles can be decoupled into two sub-models: the upper layer is the optimal power scheduling on a time scale, which mainly considers the cluster time distribution characteristics of electric vehicles, and takes the minimum fluctuation of the system equivalent load and the maximum charging and discharging benefits of electric vehicle owners as the goal, and determines the charging and discharging power of electric vehicles in different optimization periods; on the basis of the upper-layer time optimization, the lower-layer spatial scheduling model takes the feeder distribution system with the goal of minimizing the active network loss as an example to analyze the spatial distribution characteristics of charging and discharging of electric vehicles, and reasonably formulate the charging and discharging allocation scheduling plan for each electric vehicle;

[0070] The time scheduling model for orderly charging and discharging of electric vehicles optimizes the charging and discharging power values ​​in each period of the day to achieve peak shaving and valley filling of the system load. It is a non-real-time scheduling method that uses the time-optimized charging and discharging power as the connection variable between the upper-level time scheduling optimization and the lower-level spatial scheduling optimization. When performing real-time spatial allocation scheduling, the number of charging vehicles is allocated according to the charging and discharging power in each period of the daily plan to achieve a balance between the supply and demand of power.

[0071] Energy storage equipment:

[0072] Under the premise of ensuring that the dispatchable capacity of user-side energy storage and the energy storage capacity of the aggregation side meet the peak-shaving demand of the system, the PC / EC configuration of the energy storage capacity on the aggregation side is studied with the goal of maximizing the annual income of the aggregator and minimizing the peak-to-valley difference of the electricity consumption curve. The income of the aggregator is the income F1 when participating in peak-shaving, and the expenditure of the aggregator is divided into three parts: the first part is the cost of aggregating user-side energy storage F2, the second part is the cost of configuring and maintaining the energy storage on the aggregation side F3, and the third part is the charging cost of the energy storage on the aggregation side F4. Therefore, with the goal of maximizing the annual income of the aggregator and minimizing the peak-to-valley difference of the electricity consumption curve, the objective function for configuring the energy storage capacity on the aggregation side is:

[0073]

[0074]

[0075] Considering the peak-shaving power and capacity constraints, the power balance constraints of the substation, and the energy storage operation constraints, the optimization algorithm is used to obtain the optimal scheduling strategy for energy storage equipment.

[0076] Embodiment 2:

[0077] Taking the decentralized air conditioning in residential areas as an example, the existing heat dissipation model of the building-air conditioning system is an equivalent thermal parameter model, among which the third-order equivalent thermal parameter model has the highest accuracy, but the calculation complexity is also high, and it is not applicable when the accuracy requirement is not high. Therefore, the third-order model is simplified to a first-order ETP model that describes the relationship between indoor temperature and air conditioning cooling capacity, such as Figure 1As shown in the figure, the load reduction capacity of a single air conditioner in the distribution network is limited, and the air conditioning load is highly dispersed and uncertain. Therefore, it is necessary to use load aggregators to conduct regional integrated control of the air conditioning load of residents in residential areas, so that the air conditioning unit cluster can participate in the allocation of the distribution network and expand the air conditioning load control capacity;

[0078] When considering the typical flexible load optimization problem of users in the substation area, mathematical methods are often used to express it as an optimization problem containing multiple objectives and constraints, and the NSGA-II algorithm is used for calculation, such as Figure 2 As shown in the figure, the improvements are specifically in the following three aspects: 1) A fast non-dominated sorting method is proposed, and the time complexity is reduced from the original Reduced to , the sorting speed is greatly improved. Where M is the number of objective functions, and N is the number of individuals in the population; 2) The crowding comparison method is used to replace the original sharing function. By calculating and comparing the crowding degree of the population, the Pareto solution is evenly distributed, maintaining the diversity of the population; 3) The elite strategy is introduced to retain the best individuals in the competition mechanism of survival of the fittest, so that the solution result is closer to the optimal solution;

[0079] like Figure 3 As shown in the figure, the flexible load control method of connecting new energy to multi-station distribution network under multiple time scales involves non-cooperative game theory in the day-ahead dispatching strategy of distribution network station load considering multi-agent game, and the day-ahead dispatching and real-time control method of distribution network adopts multi-objective particle swarm algorithm and adaptive grid method to find the optimal value;

[0080] Based on the analysis and research of the load characteristics and regulation characteristics of flexible loads, the flexible load aggregation modeling method is studied, pointing out that the flexible load is difficult to centrally control and dispatch due to its dispersion, and using this as a breakthrough to build a flexible load aggregation model. First, a single load model is built with typical flexible loads: single electric vehicles, air conditioners, and energy storage as representatives, and then a certain number of flexible load monomers are aggregated to form a flexible load aggregate through some aggregation methods;

[0081] Under a certain market scale, the factors affecting the electricity demand of electric vehicles can be summarized into three aspects: power batteries, charging facilities, and user behavior. Among them, the power battery capacity determines the user's charging frequency. The larger the battery capacity, the lower the user's charging frequency. The charging power is also related to the battery capacity. The larger the battery capacity, the greater the charging power. The charging power is also limited by the power level of the charging facilities. At the same time, the configuration ratio of charging facilities will also affect the time distribution of charging demand. The user behavior that affects electric vehicles mainly includes the start time of charging and the daily mileage. The more concentrated the user's start charging time, the greater the charging power that the power grid needs to provide; and the daily mileage reflects the user's daily power consumption. Under a certain charging power, the mileage is related to the charging duration;

[0082] At present, most of the temperature control loads control the indoor temperature through a simple switch control strategy. The switch control strategy is a relatively simple direct load control method, which will quickly and directly reduce the total load power of the system. However, with the development and reform of the power industry, users will not be satisfied with participating in simple and direct demand response. The future demand response model should increasingly take into account the diversified electricity demand of users. Therefore, the impact of user behavior on the load group response is taken into account, and the load characteristics of the temperature control load under different user electricity behaviors are analyzed. The equivalent parameter method based on thermodynamic modeling is used to establish the air conditioning load model;

[0083] Since the power system contains generators with large power uncertainty such as wind farms, the electric energy generated by the system is often not fully absorbed. Therefore, energy storage devices play an important role in the operation and control of the power grid. The characteristics of energy storage devices that can transfer energy in time and space also make their application in power systems more and more in-depth. According to the power level and response speed of various energy storage methods, their application occasions are also different. For example, pumped storage and compressed air have large capacity and low cost, but long operation cycle, and are mainly used for peak and frequency regulation and peak shaving and valley filling of the system; while batteries have high energy density and fast response speed, and can be used to suppress power fluctuations of the system;

[0084] Aiming at the centralized control demand of loads with heterogeneous parameters and structures at the resource aggregator level, a state queue algorithm based on load controllable margin index is proposed on the basis of comprehensive consideration of load power margin, controllable time margin and controllable energy state. Figure 8 As shown, the specific steps are as follows:

[0085] Step 1: Group the loads according to their own operating status variables;

[0086] Step 2: Form a normalized controllable margin index for the loads in the controllable group, sort them according to the size of the comprehensive index, and set the action threshold of the controllable coefficient;

[0087] Step 3: Screen the loads within the action threshold range, determine the number of responses, and perform direct load control;

[0088] Each load group takes the minimum deviation of the control target as the objective function, solves the number of response loads in the group and controls them accordingly. In each period, since the load controllable margin index in each load group is time-varying, the load controlled in each period is also changing.

[0089] Embodiment three:

[0090] Optimize the dispatching of multiple distribution networks connected to new energy sources, that is, control the energy of the new energy generation units, energy storage units, and interactive units of multiple distribution networks through energy management units, so as to achieve peak load shaving and valley filling and make full use of energy, so as to reduce its operating costs and improve its comprehensive benefits;

[0091] The main market players are distribution operators and load aggregators distributed in the distribution network. Distribution operators do not have pricing and bargaining power and can only reduce operating costs by using the difference in power grid sales and purchase prices and charging network access fees. Load aggregators have independent bargaining power and can determine the external power transaction volume and transaction price through negotiation based on their own power generation and consumption needs and operating constraints, with the goal of minimizing the operating costs of load aggregators. Load aggregators can only trade with each other and provide feedback to distribution operators based on the negotiated transaction objects and corresponding transaction volume information. The negotiated transaction volume may not necessarily meet the requirements. The internal power of the load aggregator is balanced. Therefore, when the internal power is unbalanced, the load aggregator operator needs to buy or sell electricity from the distribution operator. In the case of multi-station dispatching in the distribution network, there are multiple load aggregators. According to the positive or negative difference between the internal renewable energy power and the load power of the load aggregator in the station area, some demand-side load aggregators pay the power purchase price to purchase the excess electricity of other microgrids. The higher the bid, the more energy is obtained. Multiple demand-side load aggregator microgrids form a game situation through bidding. The supply-side load aggregator provides energy to the demand-side according to the bid ratio and obtains benefits according to the proportion of energy provided by itself.

[0092] There is a flow of energy and information between the three entities involved in the regulation process: the upper grid, the distribution operator, and the load aggregator. Fig. 9 As shown;

[0093] Each demand-side load aggregator competes with other load aggregators to obtain energy provided by the supply side by changing its bid to maximize its own interests. At the same time, the microgrid needs to consider the overall power balance and the constraints of each internal unit during the bidding process;

[0094] Considering the characteristics of volatility, intermittency and uncertainty of distributed renewable energy such as wind power and photovoltaic power, if the strategy of day-ahead scheduling is planned based on the results of day-ahead prediction, and only the day-ahead scheduling scheme is implemented in intraday scheduling, the prediction error of renewable energy may endanger the safe and stable operation of the system and cause great fluctuations. The prediction accuracy of distributed renewable energy will gradually increase with the approach of time. The closer the prediction time is, the higher the prediction accuracy will be, and the smaller the system uncertainty will be. In the same situation, for flexible loads, the load prediction accuracy at different time scales is also different, and the resources of flexible loads participating in the power grid, the time of flexible load response, the speed of response and the duration of the response are also different at different time scales. Both the source and the load must be adjusted differently at different time scales. In order to reduce the error, the traditional transmission network implements day-ahead and real-time scheduling. After the distributed power source is added to the distribution network, the distribution network can autonomously control the resource scheduling within the region, which is significantly different from the traditional distribution network that only serves as a receiving end. The present invention imitates the multi-time scale scheduling of the transmission network to establish three time scales to conduct active distribution network resource scheduling research, so that it can fully tap the scheduling potential of flexible loads and smooth the influence of uncertainty on both the source and the load.

[0095] Multi-time scale dispatching of transmission network Establish three time scales to conduct dispatching research on the resources of the entire active distribution network, so that it can fully tap the dispatching potential of flexible loads and smooth the impact of uncertainties in both sources and loads, such as Fig.10 As shown, after establishing the multi-time scale scheduling framework, the particle swarm algorithm is used to solve it;

[0096] The two time scales of "day-ahead-real-time" can fully utilize the resource potential of the entire distribution network, maximize the economic benefits of the power grid, smooth out the fluctuations in renewable energy processing, and ensure the safe and stable operation of the power grid. Fig.11 As shown in the figure, the power value predicted by the renewable distributed power source and the power demand value predicted by the load are used in the day-ahead dispatching, and the optimization is carried out with multiple objectives such as the best economic benefit of the power grid, the minimum active network loss, the minimum voltage offset, and the minimum wind and solar abandonment rate. The power grid side finally determines the day-ahead dispatching plan according to its own goals and the comprehensive requirements of the upper power grid. The dispatching output of the day-ahead flexible load can be released and the load situation can be updated. In the real-time dispatching during the day, the flexible load is mainly dispatched with the highest economic benefit of the distribution network to meet the situation of the fluctuation of renewable energy processing during the day, so that the power grid and the economy can operate stably. The real-time dispatching is mainly to avoid the occurrence of emergencies, and mainly dispatches resources such as flexible loads with quick response to ensure the reliable operation of the power grid.

[0097] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An energy system optimization scheduling method considering flexible loads, comprising the following steps: S1: Utilize new technologies such as smart home appliance networking, orderly charging of electric vehicles, and household energy storage to coordinate the management of flexible loads in the substation, such as electric vehicles, household energy storage, and smart IoT appliances. Establish a load aggregation model by analyzing the load regulation characteristics. S2: Then formulate the coordinated optimization dispatch strategy of flexible load on the substation side; S3: Considering the scenario where multiple substations participate in grid dispatching, a control framework is established and dispatching strategies under multiple time scales are proposed; S4: Starting from the new background and new situation, using emerging technologies, we will tap into the flexible load resource regulation capabilities of the substation, achieve the goals of optimal power consumption in the substation, smooth out fluctuations in new energy, and ensure safe, stable and economical operation of the power grid. We will increase the new energy absorption rate, protect the vital interests of users, and provide new ideas for optimal operation management and energy efficiency improvement of future distribution network substations.

2. The energy system optimization scheduling method considering flexible load according to claim 1 is characterized by: Typical flexible loads of users in the substation include temperature control loads, electric vehicles and energy storage equipment. Since flexible loads have good adjustment characteristics, typical types of flexible loads are considered to participate in the substation optimization scheduling.

3. The energy system optimization scheduling method considering flexible load according to claim 1 is characterized in that: Temperature loads use temperature control to achieve more refined regulation. The temperature control of temperature-controlled loads is mainly based on changes in electricity prices and outdoor temperature. The load is transferred from high electricity price periods to low electricity price periods to reduce user costs, or the temperature setting value is adjusted according to the system operation conditions to provide auxiliary services for the system and ensure the stable operation of the power system.

4. The energy system optimization scheduling method considering flexible load according to claim 1 is characterized by: The load aggregator controls the temperature-controlled load by means of temperature control. Considering the benefits of the control behavior and giving users certain subsidies, the optimization goal is to maximize the interests of the load aggregator: Where G represents the interest of the load aggregator.

5. The energy system optimization scheduling method considering flexible load according to claim 1 is characterized in that: Users are often concerned about their own comfort. The comfort of temperature control load is closely related to temperature. Therefore, the user level takes the minimum temperature change as the optimization goal, which is expressed as follows: Wherein, Tk set represents the temperature setting value; T0 set represents the temperature expected value.

6. The energy system optimization scheduling method considering flexible loads according to claim 1 is characterized by: The aggregate power constraints are: Where P dr,t,min and P dr,t,max They respectively represent the minimum and maximum values ​​of the demand-side response. When the temperature changes by more than 3°C, the human body will obviously feel uncomfortable, so the temperature change needs to be constrained within 3°C. Finally, the intelligent optimization algorithm is used to obtain the temperature control load optimization scheduling strategy.

7. The energy system optimization scheduling method considering flexible load according to claim 1 is characterized by: The time-optimized charging and discharging power is used as the connecting variable for the upper-layer time scheduling optimization and the lower-layer space scheduling optimization. When performing real-time space allocation scheduling, the number of charging vehicles is allocated according to the charging and discharging power in each period of the daily plan to achieve a balance between supply and demand of power.

8. The energy system optimization scheduling method considering flexible load according to claim 1 is characterized by: On the premise of ensuring that the dispatchable capacity of user-side energy storage and the energy storage capacity of aggregation side meet the peak-shaving demand of the system, the PC / EC configuration of the energy storage capacity on the aggregation side is studied with the goal of maximizing the annual profit of the aggregator and minimizing the peak-to-valley difference of the electricity consumption curve. The profit of the aggregator is the profit F1 when participating in peak-shaving, and the expenditure of the aggregator is divided into three parts.

9. The energy system optimization scheduling method considering flexible load according to claim 1 is characterized by: The first part is the cost of aggregating user-side energy storage F2, the second part is the cost of configuring and maintaining the aggregate-side energy storage F3, and the third part is the charging cost of the aggregate-side energy storage F4. Therefore, with the goal of maximizing the annual revenue of the aggregator and minimizing the peak-to-valley difference of the electricity consumption curve, the objective function for configuring the aggregate-side energy storage capacity is: Considering the peak-shaving power and capacity constraints, the power balance constraints of the substation, and the energy storage operation constraints, the optimization algorithm is used to obtain the optimal scheduling strategy for energy storage equipment.

Images