Adjustable resource aggregation scheduling method and system for power distribution network

Abstract

The invention relates to the technical field of resource scheduling, and discloses an adjustable resource aggregation scheduling method and system for a power distribution network, and the method comprises the steps: building a building thermal inertia model, and obtaining a thermal inertia value of each building; acquiring load data and indoor temperature parameters of each building, and determining a temperature sensitive load proportion and an adjustable load of the building; establishing a load prediction model according to the historical load data to obtain a load prediction result, and performing thermal inertial sensitivity correction on the load prediction result according to the proportion of the temperature sensitive load and the time-of-use electricity price of the power distribution network; and according to the thermal inertia value, the time-of-use electricity price, the temperature sensitive load proportion and the adjustable load, calculating the charge and discharge priority of each building, and according to the charge and discharge priority and the corrected load prediction result, generating an energy storage scheduling scheme. According to the invention, through quantification of building group heterogeneity, dynamic prediction of load demands and collaborative optimization of a charging and discharging strategy, the overall demand response capability is significantly improved, and an efficient collaborative scheme is provided for intelligent power grid and building energy management.

Description

Technical Field

[0001] The present invention relates to the technical field of resource scheduling, and in particular to a method and system for aggregate scheduling of adjustable resources in a distribution network. Background Art

[0002] In modern building energy management systems, the energy collaborative optimization of building clusters is increasingly valued. However, the current traditional distribution network dispatching technology still has great limitations in the collaborative optimization of multiple buildings. The core problem is that when responding to the peak and valley electricity price signals of the distribution network, each building in the building cluster often finds it difficult to form a unified dispatching strategy, resulting in low overall energy utilization efficiency and even exacerbating grid load fluctuations. The root cause of this lack of coordination lies in the significant heterogeneity within the building cluster.

[0003] On the one hand, due to the differences in thermal inertia of each building, their response speeds to temperature changes are different, which in turn affects their energy storage potential. On the other hand, the energy storage devices equipped in each building have different capacities and charging and discharging rates, which limits their flexibility in participating in grid dispatching. In addition, due to significant differences in the load characteristics of each building, such as peak load and load fluctuation frequency, their response requirements to electricity price signals vary. The combined effect of these factors leads to the fact that when the building group responds to the peak and valley electricity price signals of the distribution network, each building only responds independently according to its own situation, lacking global coordination, resulting in poor overall load regulation. Some buildings even concentrate energy storage during the low electricity price period, resulting in new load peaks, which in turn increases the burden on the power grid.

[0004] Due to the lack of coordination, the overall adjustable load response capability of the building group is far lower than the sum of the response capabilities of each building, and it cannot fully play its role in power grid peak regulation. This not only seriously restricts the efficiency improvement of the building energy management system, but also weakens the distribution network's control ability over the adjustable load. Therefore, how to overcome the heterogeneity within the building group and achieve effective coordinated scheduling has become a key issue that needs to be solved urgently. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides an adjustable resource aggregation scheduling method and system for a distribution network, so as to overcome the heterogeneity within a building group, effectively improve the overall demand response capability of the building, and achieve effective collaborative scheduling.

[0006] In a first aspect, the present invention provides a method for aggregate scheduling of adjustable resources in a distribution network, the method comprising: Obtain the temperature distribution parameters and building material parameters of each building in the building group, and establish a building thermal inertia model based on the temperature distribution parameters and building material parameters to obtain the thermal inertia value of each building; Obtain the total load data, energy storage device data and indoor temperature parameters of each building in the building group, and determine the temperature-sensitive load proportion and adjustable load of each building based on the total load data, energy storage device data, indoor temperature parameters and thermal inertia value; According to the historical load data of the building group, a time-scale-based load forecasting model is established to obtain the load forecasting results. According to the proportion of temperature-sensitive loads and the time-of-use electricity price of the distribution network, the load forecasting results are corrected for thermal inertia sensitivity to obtain the corrected load forecasting results. According to the thermal inertia value, time-of-use electricity price, proportion of temperature-sensitive load and adjustable load, the charging and discharging priority of each building is calculated, and the energy storage scheduling plan is generated based on the charging and discharging priority and the corrected load forecast results.

[0007] Furthermore, the step of establishing a building thermal inertia model according to the temperature distribution parameters and the building material parameters to obtain the thermal inertia value of each building includes: Calculate the building heat dissipation coefficient based on the temperature distribution parameters of the building surface, the thermal conductivity of the building materials and the thickness of the building walls of each building; According to the building heat dissipation coefficient and building surface area, the building thermal inertia model is established using the heat balance equation to obtain the thermal inertia value of each building.

[0008] Furthermore, the step of determining the proportion of temperature-sensitive load and adjustable load of each building according to the total load data, indoor temperature parameters and thermal inertia value includes: According to the indoor temperature parameters, the temperature difference between the indoor target temperature and the indoor actual temperature is calculated, and the temperature sensitive load component is calculated according to the temperature difference and the thermal inertia value; Determine the proportion of temperature-sensitive loads of each building based on the ratio between the temperature-sensitive load component and the total load data; Determine the actual charge and discharge rate of the energy storage device according to the rated power of the energy storage device of each building, and determine the upper limit of the device power according to the preset response time and the actual charge and discharge rate; The current available capacity of the energy storage device is obtained, and the adjustable load is determined based on the current available capacity and the upper power limit of the device.

[0009] Furthermore, the temperature-sensitive load component is expressed by the following formula: In the formula, Q thermal represents the temperature-sensitive load component, Indicates the indoor target temperature, Indicates the actual indoor temperature, and H indicates the thermal inertia value; The following formula is used to express the proportion of temperature sensitive load: In the formula, R thermal Indicates the proportion of temperature sensitive load, L total Indicates total load data; The adjustable load is expressed by the following formula: In the formula, △Q represents the adjustable load, Indicates the current available capacity of the energy storage device, P rate Indicates the actual charge and discharge rate, Represents the response time window.

[0010] Furthermore, the step of establishing a load forecasting model based on a time scale according to the historical load data of the building group to obtain a load forecasting result includes: Divide the historical load data of the building group according to the time scale to obtain interval load data, which includes short-term load data, medium-term load data and long-term load data; According to the load data of each interval and the corresponding preset algorithm, a sub-load forecasting model at each time scale is established, and the sub-load forecasting model is weighted and summed according to the preset weights to obtain the load forecasting model; According to the load forecasting model, the load forecasting result is obtained.

[0011] Furthermore, the steps of establishing sub-load forecasting models at various time scales according to the load data of each interval and the corresponding preset algorithm, and performing weighted summation on each sub-load forecasting model according to preset weights to obtain the load forecasting model include: According to the short-term load data and autoregressive model, a short-term sub-load forecasting model is established; According to the medium-term load data and exponential smoothing model, a medium-term sub-load forecasting model is established; According to the long-term load data and grey model, a long-term sub-load forecasting model is established; Based on the rolling time window, the predicted value and actual value of each sub-load forecasting model are obtained to obtain the combined forecasting total error. The sub-load forecasting model includes a short-term sub-load forecasting model, a medium-term sub-load forecasting module and a long-term sub-load forecasting model. Taking the minimization of the total error of combined forecasting as the objective function and the sum of the weights of each sub-load forecasting model equal to the threshold as the constraint condition, a weight optimization model is established; The least square method is used to solve the weight optimization model to obtain the optimal weight of each sub-load forecasting model; The load forecasting model is obtained by weighted summing up each sub-load forecasting model according to the optimal weight.

[0012] Furthermore, the step of performing thermal inertia sensitivity correction on the load forecast result according to the proportion of temperature sensitive loads and the time-of-use electricity price of the distribution network to obtain the corrected load forecast result includes: Determine the price difference based on the time-of-use electricity price of the distribution network, and determine the price fluctuation coefficient based on the ratio of the price difference to the benchmark electricity price; According to the electricity price fluctuation coefficient and the proportion of temperature-sensitive load, the thermal inertia sensitivity correction coefficient is determined, and the thermal inertia sensitivity correction coefficient is multiplied by the load forecast result to obtain the corrected load forecast result; Among them, the thermal inertia sensitivity correction coefficient is expressed by the following formula: In the formula, δ represents the thermal inertia sensitivity correction coefficient, μ represents the price elasticity coefficient, △P represents the electricity price difference, represents the base electricity price, R thermal Indicates the proportion of temperature sensitive load.

[0013] Furthermore, the step of calculating the charging and discharging priority of each building according to the thermal inertia value, time-of-use electricity price, temperature-sensitive load proportion and adjustable load includes: Determine the price difference based on the time-of-use electricity price of the distribution network, and determine the economic priority based on the product of the price difference and the proportion of temperature-sensitive loads; Determine the response speed priority based on the thermal inertia value; Determine resource utilization priorities based on the current available capacity and adjustable load of the energy storage device; Calculate charging and discharging priorities based on economic priorities, response speed priorities, and resource utilization priorities; Among them, the following formula is used to express the charging and discharging priority: In the formula, α represents the charge and discharge priority, △P represents the electricity price difference, H represents the thermal inertia value, R thermal represents the proportion of temperature sensitive load, △Q represents the adjustable load, Indicates the current available capacity.

[0014] Furthermore, the step of generating an energy storage scheduling plan according to the charging and discharging priorities and the corrected load forecast result includes: According to the revised load forecast results, the load forecast period is divided and the charging and discharging requirements in different periods are determined; According to different time periods, the charging and discharging priorities of each building are sorted to obtain the sorting lists corresponding to different time periods; According to the charging and discharging demands and sorting lists corresponding to different time periods, an energy storage scheduling plan for the building group based on time-sharing charging and discharging power allocation is generated.

[0015] In a second aspect, the present invention provides an adjustable resource aggregation scheduling system for a distribution network, the system comprising: The load analysis module is used to obtain the temperature distribution parameters and building material parameters of each building in the building group, and to establish a building thermal inertia model based on the temperature distribution parameters and building material parameters to obtain the thermal inertia value of each building; Obtain the total load data, energy storage device data and indoor temperature parameters of each building in the building group, and determine the temperature-sensitive load proportion and adjustable load of each building based on the total load data, energy storage device data, indoor temperature parameters and thermal inertia value; The load forecasting module is used to establish a time-scale-based load forecasting model based on the historical load data of the building group to obtain the load forecasting results, and to perform thermal inertia sensitivity correction on the load forecasting results based on the proportion of temperature-sensitive loads and the time-of-use electricity price of the distribution network to obtain the corrected load forecasting results; The collaborative scheduling module is used to calculate the charging and discharging priority of each building based on the thermal inertia value, time-of-use electricity price, temperature-sensitive load ratio and adjustable load, and generate an energy storage scheduling plan based on the charging and discharging priority and the corrected load forecast results.

[0016] The present invention provides an adjustable resource aggregation scheduling method and system for a distribution network. The present invention is based on the thermal inertia value of buildings and the decomposition of temperature-sensitive loads, which can improve the accuracy of load prediction and provide data support for subsequent refined scheduling. Through a priority strategy based on electricity price differences and temperature-sensitive responses, the energy cost can be effectively reduced and the group scheduling benefits can be improved. At the same time, through the dual constraint mechanism of equipment power and capacity constraints, the safety of equipment and power grids is guaranteed and the reliability of the power grid is improved. Finally, based on a rolling optimization and real-time correction mechanism, it dynamically responds to environmental changes and improves the robustness of the scheduling scheme.

[0017] The present invention significantly improves the overall demand response capability of building groups by quantifying the heterogeneity of building groups, dynamically predicting load demand and collaboratively optimizing charging and discharging strategies, and provides a standardized, scalable and efficient collaborative solution for building groups to participate in the demand response of new power systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flow chart of an adjustable resource aggregation scheduling method for a distribution network in an embodiment of the present invention; Figure 2 It is a structural diagram of an adjustable resource aggregation scheduling system for a distribution network in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0020] See also Figure 1 The first embodiment of the present invention provides a method for aggregate scheduling of adjustable resources in a distribution network, comprising steps S10 to S40: Step S10, obtaining temperature distribution parameters and building material parameters of each building in the building group, and establishing a building thermal inertia model according to the temperature distribution parameters and building material parameters to obtain the thermal inertia value of each building; Step S20, obtaining total load data, energy storage device data and indoor temperature parameters of each building in the building group, and determining the temperature sensitive load proportion and adjustable load of each building according to the total load data, energy storage device data, indoor temperature parameters and thermal inertia value; Step S30, based on the historical load data of the building group, a load forecasting model based on a time scale is established to obtain a load forecasting result, and the load forecasting result is corrected for thermal inertia sensitivity based on the proportion of temperature-sensitive loads and the time-of-use electricity price of the distribution network to obtain a corrected load forecasting result; Step S40, calculating the charging and discharging priority of each building according to the thermal inertia value, time-of-use electricity price, temperature-sensitive load ratio and adjustable load, and generating an energy storage scheduling plan according to the charging and discharging priority and the corrected load forecast result.

[0021] The present invention provides a collaborative optimization scheduling method for adjustable resources of a building group in a distribution network, which is intended to solve the problem of low efficiency when the current building group independently responds to power grid electricity price signals due to differences in thermal inertia, differences in energy storage device capacity / response speed, and heterogeneity of load characteristics. Therefore, it is first necessary to analyze the thermal inertia of each building in the building group.

[0022] The thermal inertia of a building refers to a property of building materials or building structures in terms of absorbing, storing and releasing heat. Thermal inertia is related to the absorption, storage and release of heat. For example, building materials with high thermal inertia can absorb and store a large amount of heat when the outside temperature rises. When the outside temperature drops, these building materials that store heat will slowly release the stored heat, thereby maintaining a relatively stable indoor temperature. Thermal inertia enables buildings to buffer temperature changes. It can slow down the response speed of indoor temperature to changes in outside temperature and reduce the amplitude of indoor temperature fluctuations, thereby helping to reduce the heating and cooling energy consumption of buildings.

[0023] Since the influence of the thermal inertia of the building on the building load is not considered in the coordinated dispatching process of the building group participating in the power grid, in the dispatching method provided by the present invention, the thermal inertia of the building is first calculated, and then the influence of the thermal inertia on the load power is analyzed. In a preferred embodiment, the present invention calculates the thermal inertia value of each building in the building group by the following steps: Calculate the building heat dissipation coefficient based on the temperature distribution parameters of the building surface, the thermal conductivity of the building materials and the thickness of the building walls of each building; According to the building heat dissipation coefficient and building surface area, the building thermal inertia model is established using the heat balance equation to obtain the thermal inertia value of each building.

[0024] In this embodiment, the temperature distribution parameters of the building outer surface are collected by an infrared sensor array, the temperature change is determined according to the temperature distribution parameters, and the temperature change rate is obtained by dividing the temperature change by the unit time; then, the thermal conductivity of the building materials of each building is read from the building material parameter library, and the building heat dissipation coefficient of each building is determined according to the temperature change, the thermal conductivity of the building materials and the wall thickness of the building. The calculation formula is expressed as follows: In the formula, k represents the building heat dissipation coefficient, λ represents the thermal conductivity of building materials, △T represents the temperature change, and d represents the wall thickness.

[0025] Based on the calculated building heat dissipation coefficient and the surface area of ​​the building exterior wall, the thermal inertia model of the building is constructed using the heat balance equation to determine the thermal inertia value of each building. The thermal inertia model can be expressed as: In the formula, H represents thermal inertia, A represents building surface area, C p Represents the specific heat capacity of building materials.

[0026] This embodiment quantifies the heat storage capacity of each building, thereby providing accurate basic data for subsequent load modeling.

[0027] Based on the heat storage capacity of the building, the load characteristics of each building are analyzed below, and the temperature-sensitive load is separated from the total load of the building. At the same time, by accurately identifying the adjustable load, unnecessary scheduling is avoided, thereby improving the scheduling capability. The separation steps of the temperature-sensitive load and the specific calculation steps of the adjustable load include: According to the indoor temperature parameters, the temperature difference between the indoor target temperature and the indoor actual temperature is calculated, and the temperature sensitive load component is calculated according to the temperature difference and the thermal inertia value; Determine the proportion of temperature-sensitive loads of each building based on the ratio between the temperature-sensitive load component and the total load data; Determine the actual charge and discharge rate of the energy storage device according to the rated power of the energy storage device of each building, and determine the upper limit of the device power according to the preset response time and the actual charge and discharge rate; The current available capacity of the energy storage device is obtained, and the adjustable load is determined based on the current available capacity and the upper power limit of the device.

[0028] In this embodiment, the change of the cooling / heating load of the building is directly related to the thermal inertia of the building and the temperature difference between the inside and outside, which conforms to the thermal balance equation. Therefore, by obtaining the actual indoor temperature and the indoor target temperature of the building, calculating the temperature difference between the two temperatures, and multiplying the temperature difference with the thermal inertia calculated in the above steps, the temperature-sensitive load component can be separated from the total load. The formula is expressed as follows: In the formula, Q thermal represents the temperature-sensitive load component, Indicates the indoor target temperature, represents the actual indoor temperature, and H represents the thermal inertia value.

[0029] Then, the total load data of the building is obtained, and the temperature-sensitive load component is divided by the total load data to obtain the temperature-sensitive load ratio, which is expressed as follows: In the formula, R thermal Indicates the proportion of temperature sensitive load, L total Indicates total load data.

[0030] In this embodiment, in addition to separating the temperature-sensitive load components, the adjustable load is also identified, wherein the adjustable load is used to quantify the maximum adjustable load of the building energy storage device within a specific time window, and its formula is expressed as follows: In the formula, △Q represents the adjustable load, Indicates the current available capacity of the energy storage device, P rate Indicates the actual charge and discharge rate, Represents the response time window.

[0031] Specifically, the current available capacity of the energy storage device refers to the energy that has been stored but not released by the energy storage device at the current moment, that is, the remaining capacity of the energy storage device, and its value can be obtained by subtracting the used capacity from the rated capacity of the energy storage device; the actual charge and discharge rate refers to the maximum charge and discharge power under the current working conditions, and its value can be obtained by multiplying the rated power of the energy storage device by the equipment efficiency coefficient; the response time window refers to the time period in which the power grid requires the building to complete load regulation. The time window is determined by the dispatch instruction, and the corresponding response time window varies according to different scenarios and uses. For example, in the frequency modulation scenario, in order to quickly smooth the high-frequency fluctuations of the power grid, the time window issued by the power grid is 0.5 hours. In the peak shaving and valley filling scenario, in order to reduce the peak-to-valley difference of the power grid, the time window issued is 2 hours. The product of the actual charge and discharge rate and the response time window represents the upper limit of the device power. Therefore, in this embodiment, the adjustable load is the minimum value of the current available capacity and the upper limit of the device power.

[0032] This embodiment uses capacity constraints to ensure that the dispatch amount does not exceed the current available capacity of the energy storage device to avoid equipment shutdown due to excessive discharge. Power constraints prevent the power limit of the equipment from being exceeded to avoid efficiency degradation or hardware damage due to overload. This embodiment uses a dual constraint mechanism to determine the actual adjustable capacity, thereby ensuring the feasibility of subsequent dispatch strategies.

[0033] On the basis of the above steps, a load forecasting model for a building group is established based on the historical load data of the building group. The load forecasting model can be constructed based on a neural network or data fitting method. However, in actual application scenarios, the fluctuation characteristics of load data at different time scales are not the same. Therefore, in order to improve the accuracy of the prediction results, in a preferred embodiment, the present invention adopts a time-sharing prediction method, and a combined prediction result is obtained by performing time-sharing prediction on the loads in different time periods. The specific steps include: Divide the historical load data of the building group according to the time scale to obtain interval load data, which includes short-term load data, medium-term load data and long-term load data; According to the load data of each interval and the corresponding preset algorithm, a sub-load forecasting model at each time scale is established, and the sub-load forecasting model is weighted and summed according to the preset weights to obtain the load forecasting model; According to the load forecasting model, the load forecasting result is obtained.

[0034] In this embodiment, the historical load data is first divided based on different time scales. Preferably, the time scale is set to daily scale, weekly scale and monthly scale. According to different time scales, the cluster analysis method is used to classify the historical load data respectively, so as to obtain short-term load data, medium-term load data and long-term load data, which correspond to daily scale load data, weekly scale load data and monthly scale load data respectively.

[0035] The load fluctuation characteristics reflected by the load data at different time scales are also different. For example, the intraday fluctuation is mainly manifested as two peaks in the morning and evening, the weekly fluctuation is reflected in the load difference between weekdays and weekends, and the monthly fluctuation reflects seasonal changes. Therefore, in this embodiment, for load data at different time scales, different prediction algorithms are used to establish a sub-load prediction model. Specifically, an autoregressive model is used to model short-term load data, an exponential smoothing model is used to smooth medium-term load data, and a gray model is used to process the data sparsity of long-term load data, so as to obtain sub-load prediction models at different time scales. It should be noted that the modeling algorithm used in this embodiment is only preferred and not specifically limited. Other algorithms can also be used to establish a sub-load prediction model. There are no excessive restrictions here. The specific modeling process can refer to the modeling steps of the conventional modeling algorithm, and it will not be repeated here.

[0036] Based on each sub-load forecasting model under different time scales, weighted summation is performed according to the weights of different models, so as to obtain a load forecasting model combining different time scales, wherein each weight of the sub-model can be determined by a preset value.

[0037] In a preferred embodiment, in order to improve the accuracy of the load forecasting model, the present invention provides a weight determination method based on the forecast error. In this embodiment, the forecast error of each sub-load forecasting model is first determined based on the rolling time window. Specifically, the window length and rolling step of the rolling time window are set, that is, a fixed time span (such as the past 7 days, 24 hours, etc.) is selected, and each time the window moves forward by one time unit (such as 1 day or 1 hour), the data in the window is updated. Then, according to the actual load value sequence in the time window and the forecast value sequence of each sub-model, the forecast error of each sub-model is calculated. Here, the square error is used to characterize the forecast error, and an error matrix is ​​constructed.

[0038] The weights of each sub-load forecasting model are taken as unknown quantities, and the predicted values ​​of each sub-load forecasting model are weighted and summed to obtain the combined predicted value. Then, the prediction error between the actual load value and the combined load value is minimized, that is, the square error is minimized as the objective function, and the sum of the weights of each sub-load forecasting model is equal to 1 as the constraint condition to construct a weight optimization model. Finally, the least squares method is used to solve the weight optimization model, that is, the Lagrange multiplier method is used, the Lagrange multiplier is introduced into the objective function, the Lagrange function is constructed, and the equation group is derived. Finally, the linear equation group is solved through the error matrix to obtain the optimal weights of each sub-load forecasting model.

[0039] After determining the optimal weights of each sub-load prediction model based on the above method, the sub-load prediction models can be weighted and summed based on the weight values ​​to obtain the load prediction model. Furthermore, the weights in this embodiment can be dynamically adjusted according to the latest errors to adapt to the changes in load to ensure the accuracy of the load prediction model.

[0040] In a preferred embodiment, in order to further improve the accuracy of the load forecasting results, the present invention also provides a method for correcting the thermal inertia sensitivity of the load forecasting results. The specific correction steps include: Determine the price difference based on the time-of-use electricity price of the distribution network, and determine the price fluctuation coefficient based on the ratio of the price difference to the benchmark electricity price; According to the electricity price fluctuation coefficient and the proportion of temperature-sensitive load, the thermal inertia sensitivity correction coefficient is determined, and the thermal inertia sensitivity correction coefficient is multiplied by the load forecast result to obtain the corrected load forecast result.

[0041] Based on the above description, it can be seen that the thermal inertia of the building and the fluctuation of the power grid electricity price will also affect the load demand of the building. In the above embodiment, the load prediction model only considers the historical load data, and does not consider the impact of the electricity price signal and the thermal inertia value on the building load, resulting in the prediction not reflecting the dynamic impact of the building's thermal characteristics. Therefore, in this embodiment, the fluctuation of electricity prices and the thermal inertia of the building are added to the consideration of load prediction.

[0042] First, according to the time-of-use electricity price of the power grid, determine the price difference between peak and valley periods, and determine the price fluctuation coefficient based on the ratio between the price difference and the benchmark electricity price. The benchmark electricity price refers to the non-time-of-use electricity price. Then, according to the price fluctuation coefficient and the proportion of temperature-sensitive loads, determine the thermal inertia sensitivity correction coefficient: In the formula, δ represents the thermal inertia sensitivity correction coefficient, μ represents the price elasticity coefficient, which ranges from 0.2 to 0.5, and △P represents the electricity price difference. represents the base electricity price, R thermalIndicates the proportion of temperature sensitive load.

[0043] The load forecast result is corrected based on the above thermal inertia sensitivity correction coefficient, so that the corrected load forecast result is: Where L` represents the load forecast result after correction, and L represents the load forecast result before correction.

[0044] In the correction logic of this embodiment, the higher the thermal inertia of the building, the greater the impact of electricity price fluctuations on load forecasting. Through the correction method in this embodiment, electricity price fluctuations and the thermal inertia of the building are taken into consideration in load forecasting, further improving the accuracy of load forecasting.

[0045] After obtaining the load forecast results of the building group, an energy storage scheduling scheme can be generated based on the load demand predicted by each building. In the present invention, the scheduling cost is effectively reduced and the scheduling efficiency is improved through the aggregate scheduling of the adjustable resources of the distribution network. When performing resource scheduling, in order to further optimize resource allocation and avoid inefficient scheduling, in a preferred embodiment, the present invention also provides a method for scheduling adjustable aggregate resources based on the charging and discharging priority of the building, wherein the step of determining the charging and discharging priority includes: Determine economic priorities based on time-of-use electricity prices in the distribution network; Determine the priority of temperature change response based on thermal inertia and the proportion of temperature-sensitive loads; Determine resource utilization priorities based on the current available capacity and adjustable load of the energy storage device; The charging and discharging priority is calculated based on the economic priority, temperature change response priority and resource utilization priority.

[0046] In this embodiment, the charging and discharging priority is dynamically evaluated for each building from multiple dimensions such as economy, temperature change response speed and resource utilization rate, so as to determine the charging and discharging priority of each building. Specifically, the economy of scheduling is characterized by the price difference, that is, the difference between peak and valley electricity prices, and the price difference should be positively correlated with the priority, that is, the larger the price difference, the higher the regulation benefit and the higher the priority should be; the sensitivity of the building to temperature changes is reflected by the proportion of temperature-sensitive loads, and the proportion of temperature-sensitive loads is positively correlated with the priority, that is, the higher the proportion of temperature-sensitive loads, the stronger the load adjustability of the building and the higher the priority; thermal inertia is used to reflect The speed at which the building responds to temperature changes is reflected in the thermal inertia and the priority is negatively correlated, that is, the greater the thermal inertia, the slower the building responds to temperature. In order to avoid scheduling delays, the priority should be lowered; the adjustable load of the energy storage device represents the scheduling potential of the building, and the adjustable load is positively correlated with the priority. The larger the adjustable load, the greater the scheduling potential and the higher the priority; the current available capacity of the energy storage device, that is, the remaining capacity of the energy storage device, is used to limit the maximum adjustable amount of the building. Therefore, the current available capacity and the priority should be negatively correlated, that is, the smaller the remaining capacity, the lower the scheduling margin. In order to prevent capacity exhaustion, the priority should be lowered.

[0047] Based on the above correlation analysis, it can be seen that the dispatching in the high electricity price difference period is significantly beneficial, and discharge should be arranged first. At the same time, the load fluctuation of buildings with a high proportion of temperature-sensitive loads is strongly related to the electricity price signal, and the cost can be significantly reduced by adjustment. Therefore, this embodiment uses the product of the electricity price difference and the proportion of temperature-sensitive loads as the economic priority, and improves the economy of dispatching through the collaborative logic of the electricity price difference and the proportion of temperature-sensitive loads. Take the thermal inertia value as the response speed priority. When the high thermal inertia value is greater than 120, the building has a large thermal inertia and a slow response speed, and it takes longer to reach the target temperature. Therefore, its priority should be lowered to avoid delays that affect the real-time regulation of the power grid. Finally, the ratio of adjustable load to current available capacity is used as the resource utilization priority. The adjustable load is used as the numerator. The larger the current adjustable capacity, the higher the dispatching flexibility. The current available capacity is used as the denominator. The smaller the remaining capacity, the fewer the remaining resources after dispatching, and the more careful allocation is required. Taking the ratio of the two as the resource utilization priority can dynamically balance the current available resources and long-term dispatching potential to prevent local resource depletion.

[0048] Based on the above priorities, in this embodiment, the charging and discharging priorities can be expressed as: In the formula, α represents the charge and discharge priority, △P represents the electricity price difference, H represents the thermal inertia value, R thermal represents the proportion of temperature sensitive load, △Q represents the adjustable load, Indicates the current available capacity. When it is less than 10% of the rated capacity, the priority is forced to be downgraded to avoid overcharging or over-discharging.

[0049] The charging and discharging priority formula in this embodiment realizes the refined scheduling of energy storage resources of the building group through the parameter coupling of economic weight, thermal inertia constraint, and resource utilization balance. It can effectively adapt to the changes in grid demand by driving the dynamic adjustment of priority through real-time electricity price signals and load forecast data, avoid response delays through thermal inertia constraints, and prevent overcharging / over-discharging through remaining capacity constraints, providing a safety guarantee for the stability of charging and discharging. This embodiment combines the priority of buildings with energy storage scheduling to achieve group collaborative optimization, thereby improving the group benefits of resource scheduling.

[0050] After obtaining the charging and discharging priorities of each building and the load forecast results of the building group based on the above steps, a distributed algorithm can be used to coordinate the charging and discharging sequence and generate an energy storage scheduling method. The specific steps include: According to the revised load forecast results, the load forecast period is divided and the charging and discharging requirements in different periods are determined; According to different time periods, the charging and discharging priorities of each building are sorted to obtain the sorting lists corresponding to different time periods; According to the charging and discharging demands and sorting lists corresponding to different time periods, an energy storage scheduling plan for the building group based on time-sharing charging and discharging power allocation is generated.

[0051] In this embodiment, first, according to the load forecast results, the charging and discharging time period is divided into peak period (high load), valley period (low load), and normal period. For example, the load greater than 800kW is regarded as the peak period, the load less than 400kW is regarded as the valley period, and the rest is regarded as the normal period. Then, based on different time periods, the charging and discharging demand is determined, such as giving priority to discharge in the peak period to reduce the load on the power grid; giving priority to charging in the valley period to use low-priced electricity; and making flexible adjustments in the normal period to maintain the energy storage state.

[0052] After time division and demand determination, dynamic priority sorting is performed by time period. That is, based on the priority matrix composed of the priorities of each building, the priorities of all buildings in each time period are extracted, and then priority sorting is performed by time period. The sorting rules are as follows: in peak periods, the buildings are arranged in descending order of priority, and high-priority buildings are discharged first; in valley periods, the buildings are arranged in ascending order of priority, and low-priority buildings are charged first.

[0053] Then, based on the sorted list, adjustable load and real-time constraints, the charging and discharging power is allocated. The allocation principle is: during peak discharge, the discharge power is allocated from high to low until the grid load reduction target is met or the building's adjustable capacity is exhausted; during valley charging, the charging power is allocated from low to high until the energy storage capacity is full or the low-price period ends.

[0054] Based on the above allocation principles, the steps of the power allocation algorithm adopted in this embodiment are as follows: first, determine the demand in different time periods: in the peak period, the load needs to be reduced, and the reduction value is the predicted load minus the grid safety threshold; in the valley period, charging is required, and the chargeable capacity is equal to the charging power that the grid can withstand.

[0055] Then, power is allocated building by building according to the sorting list. When discharging power in the peak period, the current total reduction amount and the corresponding priority ranking are first determined. According to the priority sorting sequence, the available power of each building is calculated. The available power is the smaller value of the adjustable load of the building and the upper limit of the charge and discharge rate. For example, if the adjustable load of the building is 250kW and the upper limit of the charge and discharge rate is 200kW, then its available power is 200kW. Then determine the current remaining amount to be allocated, which is the difference between the total reduction amount and the total allocation amount. The smaller value between the remaining amount to be allocated and the available power is used as the allocation amount of the building, and the current allocation is added to the total allocation amount, and the discharge power of the building in the current period is recorded. Finally, determine whether the discharge is terminated, that is, if the total allocation amount is greater than or equal to the total reduction amount, the allocation process is terminated immediately.

[0056] When allocating charging power in the valley section, first determine the current rechargeable capacity and the corresponding priority ranking. According to the priority ranking sequence, calculate the rechargeable capacity of each building. The rechargeable capacity is the difference between the rated capacity of the building and the current available capacity. For example, if the rated capacity of the building is 2000kW and the current available capacity, i.e., the remaining capacity, is 800kW, then its rechargeable capacity is 1200kW. Then determine the available power of the building. The real-time charging power of the building is subject to two restrictions: one is the restriction of the rated charging rate, and the other is the constraint of the time window, i.e., the ratio between the rechargeable capacity and the valley section duration. The smaller value of the two is taken as the available power of the building. For example, if the rechargeable capacity is 1200kW and the valley section duration is 4 hours, then the theoretical power limit is 1200 / 4=300kW, but the rated power of the equipment is 200kW, so the actual available power is 200kW. Then, based on the current remaining allocatable power of the power grid (rechargeable capacity minus total allocation), the actual allocated power of the building is determined. The actual allocated power takes the smaller value of the available power and the remaining allocatable power. Finally, the total allocation, the charging power of the building and the remaining capacity of the building are updated and recorded, and it is determined whether the termination conditions are met. If the total allocation is greater than the rechargeable capacity, the allocation process is terminated immediately.

[0057] Finally, based on the above power allocation, a time-sharing charging and discharging power allocation table is generated, and the power allocation table is rolled optimized and corrected in real time according to real-time electricity price fluctuations, load forecast errors, equipment status changes, etc., so as to realize the coordinated energy storage scheduling of the building group. The power allocation algorithm of this embodiment, during the discharge scheduling process, prioritizes high-priority buildings for scheduling, achieves maximum benefits, calculates the remaining amount to be allocated in real time, flexibly adjusts the allocated power of a single building, strictly follows the power and capacity constraints of the equipment, avoids the risk of overload, and stops allocation immediately after the goal is achieved, effectively reducing invalid operations, thereby meeting the needs of the power grid while taking into account the safety and economy of building operation. During the charging scheduling process, charging is maximized during the valley period with low electricity prices, effectively reducing energy costs, giving priority to filling buildings with low remaining capacity, improving the utilization rate of energy storage devices, calculating the remaining allocable power in real time, flexibly adjusting the allocation of a single building, strictly following the power and capacity constraints, preventing equipment overload or capacity overflow, and providing safety boundary guarantees, so that charging tasks can be completed efficiently and sufficient energy reserves can be provided for peak demand response. The energy storage scheduling scheme provided in this embodiment realizes a closed loop of the entire process from priority calculation to power allocation, thereby ensuring the efficiency and reliability of the coordinated scheduling of building groups.

[0058] This embodiment provides a method for aggregate scheduling of adjustable resources in a distribution network. The present invention improves the accuracy of load forecasting based on the thermal inertia of buildings and temperature-sensitive load decomposition, provides data support for subsequent refined scheduling, and effectively reduces energy costs and improves group scheduling benefits through priority strategies based on electricity price differences and temperature-sensitive responses. At the same time, through the dual constraint mechanism of equipment power and capacity constraints, the safety of equipment and power grids is guaranteed, overload risks are avoided, and the reliability of power grids is improved. Finally, based on rolling optimization and real-time correction mechanisms, it dynamically responds to environmental changes and improves the robustness of scheduling schemes. The present invention significantly improves the overall demand response capability of building groups by quantifying the heterogeneity of building groups, dynamically predicting load demand, and collaboratively optimizing charging and discharging strategies, and provides a standardized, scalable, and efficient collaborative solution for building groups to participate in the demand response of new power systems.

[0059] See also Figure 2 Based on the same inventive concept, a second embodiment of the present invention proposes an adjustable resource aggregation scheduling system for a distribution network, comprising: The load analysis module 10 is used to obtain the temperature distribution parameters and building material parameters of each building in the building group, and to establish a building thermal inertia model based on the temperature distribution parameters and building material parameters to obtain the thermal inertia value of each building; Obtain the total load data, energy storage device data and indoor temperature parameters of each building in the building group, and determine the temperature-sensitive load proportion and adjustable load of each building based on the total load data, energy storage device data, indoor temperature parameters and thermal inertia value; The load forecasting module 20 is used to establish a load forecasting model based on a time scale according to the historical load data of the building group, obtain a load forecasting result, and perform thermal inertia sensitivity correction on the load forecasting result according to the proportion of temperature-sensitive loads and the time-of-use electricity price of the distribution network to obtain a corrected load forecasting result; The collaborative scheduling module 30 is used to calculate the charging and discharging priority of each building according to the thermal inertia value, time-of-use electricity price, temperature-sensitive load ratio and adjustable load, and generate an energy storage scheduling plan according to the charging and discharging priority and the corrected load forecast results.

[0060] The technical features and technical effects of the adjustable resource aggregation scheduling system for the distribution network proposed in the embodiment of the present invention are the same as the method proposed in the embodiment of the present invention, and will not be repeated here. Each module in the above-mentioned adjustable resource aggregation scheduling system for the distribution network can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0061] In summary, an embodiment of the present invention proposes a method and system for aggregate scheduling of adjustable resources in a distribution network. The method obtains temperature distribution parameters and building material parameters of each building in a building group, and establishes a building thermal inertia model based on the temperature distribution parameters and building material parameters to obtain the thermal inertia value of each building; obtains total load data, energy storage device data and indoor temperature parameters of each building in the building group, and determines the temperature-sensitive load proportion and adjustable load of each building based on the total load data, energy storage device data, indoor temperature parameters and thermal inertia value; establishes a load forecasting model based on a time scale based on the historical load data of the building group to obtain a load forecasting result, and performs thermal inertia sensitivity correction on the load forecasting result based on the temperature-sensitive load proportion and the time-of-use electricity price of the distribution network to obtain a corrected load forecasting result; calculates the charging and discharging priority of each building based on the thermal inertia value, time-of-use electricity price, temperature-sensitive load proportion and adjustable load, and generates an energy storage scheduling plan based on the charging and discharging priority and the corrected load forecasting result. The present invention improves the accuracy of load forecasting based on the thermal inertia value and temperature-sensitive load decomposition of buildings, provides data support for subsequent refined scheduling, and effectively reduces energy costs and improves group scheduling benefits through priority strategies based on electricity price differences and temperature-sensitive responses. At the same time, through the dual constraint mechanism of equipment power and capacity constraints, it ensures the safety of equipment and power grids, avoids overload risks, and improves the reliability of the power grid. Finally, based on rolling optimization and real-time correction mechanisms, it dynamically responds to environmental changes and improves the robustness of scheduling schemes. The present invention significantly improves the overall demand response capability of building groups by quantifying the heterogeneity of building groups, dynamically predicting load demand, and collaboratively optimizing charging and discharging strategies, and provides a standardized, scalable, and efficient collaborative solution for building groups to participate in the demand response of new power systems.

[0062] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0063] The above-mentioned embodiments only express several preferred implementation modes of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in the technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be based on the protection scope of the claims.

Claims

1. A method for aggregate scheduling of adjustable resources in a distribution network, characterized in that: include: Obtain the temperature distribution parameters and building material parameters of each building in the building group, and establish a building thermal inertia model based on the temperature distribution parameters and building material parameters to obtain the thermal inertia value of each building; Obtain the total load data, energy storage device data and indoor temperature parameters of each building in the building group, and determine the temperature-sensitive load proportion and adjustable load of each building based on the total load data, energy storage device data, indoor temperature parameters and thermal inertia value; According to the historical load data of the building group, a time-scale-based load forecasting model is established to obtain the load forecasting results. According to the proportion of temperature-sensitive loads and the time-of-use electricity price of the distribution network, the load forecasting results are corrected for thermal inertia sensitivity to obtain the corrected load forecasting results. According to the thermal inertia value, time-of-use electricity price, proportion of temperature-sensitive load and adjustable load, the charging and discharging priority of each building is calculated, and the energy storage scheduling plan is generated based on the charging and discharging priority and the corrected load forecast results.

2. The adjustable resource aggregation scheduling method of the distribution network according to claim 1 is characterized in that: The step of establishing a building thermal inertia model according to the temperature distribution parameters and the building material parameters to obtain the thermal inertia value of each building comprises: Calculate the building heat dissipation coefficient based on the temperature distribution parameters of the building surface, the thermal conductivity of the building materials and the thickness of the building walls of each building; According to the building heat dissipation coefficient and building surface area, the building thermal inertia model is established using the heat balance equation to obtain the thermal inertia value of each building.

3. The adjustable resource aggregation scheduling method of the distribution network according to claim 1 is characterized in that: The step of determining the temperature sensitive load proportion and adjustable load of each building according to the total load data, indoor temperature parameters and thermal inertia value comprises: According to the indoor temperature parameters, the temperature difference between the indoor target temperature and the indoor actual temperature is calculated, and the temperature sensitive load component is calculated according to the temperature difference and the thermal inertia value; Determine the proportion of temperature-sensitive loads of each building based on the ratio between the temperature-sensitive load component and the total load data; Determine the actual charge and discharge rate of the energy storage device according to the rated power of the energy storage device of each building, and determine the upper limit of the device power according to the preset response time and the actual charge and discharge rate; The current available capacity of the energy storage device is obtained, and the adjustable load is determined based on the current available capacity and the upper power limit of the device.

4. The adjustable resource aggregation scheduling method of the distribution network according to claim 3 is characterized in that: The temperature sensitive load component is expressed using the following formula: In the formula, Q thermal represents the temperature-sensitive load component, Indicates the indoor target temperature, Indicates the actual indoor temperature, and H indicates the thermal inertia value; The following formula is used to express the proportion of temperature sensitive load: In the formula, R thermal Indicates the proportion of temperature sensitive load, L total Indicates total load data; The adjustable load is expressed by the following formula: In the formula, △Q represents the adjustable load, Indicates the current available capacity of the energy storage device, P rate Indicates the actual charge and discharge rate, Represents the response time window.

5. The adjustable resource aggregation scheduling method of the distribution network according to claim 1 is characterized in that: The step of establishing a load forecasting model based on a time scale according to the historical load data of the building group to obtain the load forecasting result comprises: Divide the historical load data of the building group according to the time scale to obtain interval load data, which includes short-term load data, medium-term load data and long-term load data; According to the load data of each interval and the corresponding preset algorithm, a sub-load forecasting model at each time scale is established, and the sub-load forecasting model is weighted and summed according to the preset weights to obtain the load forecasting model; According to the load forecasting model, the load forecasting result is obtained.

6. The adjustable resource aggregation scheduling method of the distribution network according to claim 5 is characterized in that: The steps of establishing sub-load forecasting models at various time scales according to the load data of each interval and the corresponding preset algorithm, and performing weighted summation on each sub-load forecasting model according to preset weights to obtain the load forecasting model include: According to the short-term load data and autoregressive model, a short-term sub-load forecasting model is established; According to the medium-term load data and exponential smoothing model, a medium-term sub-load forecasting model is established; According to the long-term load data and grey model, a long-term sub-load forecasting model is established; Based on the rolling time window, the predicted value and actual value of each sub-load forecasting model are obtained to obtain the combined forecasting total error. The sub-load forecasting model includes a short-term sub-load forecasting model, a medium-term sub-load forecasting module and a long-term sub-load forecasting model. Taking the minimization of the total error of combined forecasting as the objective function and the sum of the weights of each sub-load forecasting model equal to the threshold as the constraint condition, a weight optimization model is established; The least square method is used to solve the weight optimization model to obtain the optimal weight of each sub-load forecasting model; The load forecasting model is obtained by weighted summing up each sub-load forecasting model according to the optimal weight.

7. The adjustable resource aggregation scheduling method of the distribution network according to claim 1 is characterized in that: The step of performing thermal inertia sensitivity correction on the load forecast result according to the proportion of temperature sensitive loads and the time-of-use electricity price of the distribution network to obtain the corrected load forecast result includes: Determine the price difference based on the time-of-use electricity price of the distribution network, and determine the price fluctuation coefficient based on the ratio of the price difference to the benchmark electricity price; According to the electricity price fluctuation coefficient and the proportion of temperature-sensitive load, the thermal inertia sensitivity correction coefficient is determined, and the thermal inertia sensitivity correction coefficient is multiplied by the load forecast result to obtain the corrected load forecast result; Among them, the thermal inertia sensitivity correction coefficient is expressed by the following formula: In the formula, δ represents the thermal inertia sensitivity correction coefficient, μ represents the price elasticity coefficient, △P represents the electricity price difference, represents the base electricity price, R thermal Indicates the proportion of temperature sensitive load.

8. The adjustable resource aggregation scheduling method of the distribution network according to claim 3 is characterized in that: The step of calculating the charging and discharging priority of each building according to the thermal inertia value, time-of-use electricity price, temperature-sensitive load proportion and adjustable load includes: Determine the price difference based on the time-of-use electricity price of the distribution network, and determine the economic priority based on the product of the price difference and the proportion of temperature-sensitive loads; Determine the response speed priority based on the thermal inertia value; Determine resource utilization priorities based on the current available capacity and adjustable load of the energy storage device; Calculate charging and discharging priorities based on economic priorities, response speed priorities, and resource utilization priorities; Among them, the following formula is used to express the charging and discharging priority: In the formula, α represents the charge and discharge priority, △P represents the electricity price difference, H represents the thermal inertia value, R thermal represents the proportion of temperature sensitive load, △Q represents the adjustable load, Indicates the current available capacity.

9. The adjustable resource aggregation scheduling method of the distribution network according to claim 1, characterized in that: The step of generating an energy storage scheduling plan according to the charging and discharging priority and the corrected load forecast result includes: According to the revised load forecast results, the load forecast period is divided and the charging and discharging requirements in different periods are determined; According to different time periods, the charging and discharging priorities of each building are sorted to obtain the sorting lists corresponding to different time periods; According to the charging and discharging demands and sorting lists corresponding to different time periods, an energy storage scheduling plan for the building group based on time-sharing charging and discharging power allocation is generated.

10. An adjustable resource aggregation dispatching system for a distribution network, characterized in that: include: The load analysis module is used to obtain the temperature distribution parameters and building material parameters of each building in the building group, and to establish a building thermal inertia model based on the temperature distribution parameters and building material parameters to obtain the thermal inertia value of each building; Obtain the total load data, energy storage device data and indoor temperature parameters of each building in the building group, and determine the temperature-sensitive load proportion and adjustable load of each building based on the total load data, energy storage device data, indoor temperature parameters and thermal inertia value; The load forecasting module is used to establish a time-scale-based load forecasting model based on the historical load data of the building group to obtain the load forecasting results, and to perform thermal inertia sensitivity correction on the load forecasting results based on the proportion of temperature-sensitive loads and the time-of-use electricity price of the distribution network to obtain the corrected load forecasting results; The collaborative scheduling module is used to calculate the charging and discharging priority of each building based on the thermal inertia value, time-of-use electricity price, temperature-sensitive load ratio and adjustable load, and generate an energy storage scheduling plan based on the charging and discharging priority and the corrected load forecast results.

Images