Trusted demand response potential evaluation method and device, electronic equipment and storage medium

Abstract

The invention provides a trusted demand response potential assessment method and device, electronic equipment and a storage medium, and relates to the technical field of electric power. According to the method, after a plurality of data sets corresponding to a target public building are obtained, a plurality of preset sub-models are trained through a training set in the data sets, and a correction set and a test set in the data sets are respectively input into the trained preset sub-models so as to obtain an upper bound and a lower bound of an energy consumption baseline corresponding to the target public building, and the upper bound and the lower bound are used for combining a preset thermal parameter algorithm. And finally, based on the upper and lower bounds of the energy consumption base line, the upper and lower bounds of the indoor thermal resistance, the comfort constraints related to the temperature and a preset evaluation algorithm, the credible demand response potential of the target public building is obtained in the form of the upper and lower bounds. Therefore, the deviation between the demand response potential evaluation result based on the grey box model and the actual response is reduced, and the benefit of the user participating in the demand response and the reliability of power system scheduling are improved.

Description

Technical Field

[0001] This application relates to the field of power technology, and in particular to a method, apparatus, electronic device and storage medium for assessing reliable demand response potential. Background Technology

[0002] Demand response refers to the technical means of guiding electricity users to adjust their electricity consumption behavior through electricity pricing or incentives to alleviate the imbalance between supply and demand in the power system. In public buildings, due to the high energy consumption and controllability of their air conditioning systems, they become an important target for demand response. Credible demand response potential (CDRP) refers to the range of response capabilities a building can provide at a certain confidence level after considering multiple uncertainties; it is a key indicator for assessing users' actual adjustability.

[0003] Current demand response potential assessment schemes are gradually replacing white-box and black-box models with gray-box models. White-box models are models built based on physical equations, which are highly interpretable but rely on precise parameters; black-box models are models trained on big data, which are highly adaptable but poorly interpretable; while gray-box models combine the advantages of both black-box and white-box models, possessing a certain degree of interpretability and adaptability.

[0004] However, while gray-box models achieve a certain balance between interpretability and accuracy, existing gray-box models mostly employ deterministic outputs, i.e., outputting a single potential value, making it difficult to quantify the impact of multiple uncertainties such as weather changes, population movement, and temporal evolution on demand response potential. This leads to a significant deviation between the demand response potential assessment results under current demand response potential assessment schemes and the actual response, affecting the benefits for users participating in response activities and the reliability of system scheduling. Summary of the Invention

[0005] The main objective of this application is to propose a reliable demand response potential assessment method, apparatus, electronic device, and storage medium, aiming to reduce the deviation between the demand response potential assessment results based on the gray box model and the actual response, thereby improving the benefits of user participation in demand response and the reliability of power system dispatch.

[0006] In a first aspect, the present invention provides a method for assessing credible demand response potential, comprising: Obtain multiple datasets corresponding to the target public building, wherein the multiple datasets include: a training set, a calibration set, and a test set. The training set and the test set each include multiple sets of energy consumption factor data and historical energy consumption data corresponding to each set of energy consumption factor data. The calibration set is the relative complement of the training set. Multiple preset sub-models are trained using the training set to obtain multiple trained preset sub-models. Based on the calibration set, the test set, and the trained preset sub-model, obtain the lower bound and upper bound of the energy consumption baseline corresponding to the target public building; Based on the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, and the preset thermal parameter algorithm, the lower boundary of the indoor thermal resistance corresponding to the lower boundary of the energy consumption baseline and the upper boundary of the indoor thermal resistance corresponding to the upper boundary of the energy consumption baseline are calculated and obtained respectively. Based on the preset temperature deviation range, the lower limit of the energy consumption baseline, the upper limit of the energy consumption baseline, the lower limit of the indoor thermal resistance, the upper limit of the indoor thermal resistance, and the preset evaluation algorithm, the range of credible demand response potential corresponding to the target public building is calculated and obtained.

[0007] In an optional implementation, the energy consumption factor data includes: time data, weather data, and personnel flow data; Before obtaining multiple datasets corresponding to the target public building, the method further includes: Obtain multiple sets of energy consumption factor data corresponding to the target public building and the historical energy consumption data corresponding to each set of energy consumption factor data; The acquisition of multiple datasets corresponding to the target public building includes: Based on the time data in each set of energy consumption factor data, a training set and a calibration set including time-series features are constructed.

[0008] In an optional implementation, the step of training multiple preset sub-models using the training set to obtain multiple trained preset sub-models includes: The training set is divided into multiple training subsets, and each training subset corresponds to a preset sub-model; The hyperparameters of the preset sub-model corresponding to each training subset are updated using the training subset and the preset optimization algorithm, until the hyperparameters of the preset sub-model reach the optimization target of the preset optimization algorithm, thereby obtaining multiple trained preset sub-models.

[0009] In an optional implementation, obtaining the lower bound and upper bound of the energy consumption baseline corresponding to the target public building based on the calibration set, the test set, and the trained preset sub-model includes: The calibration set is divided into multiple calibration subsets, and each calibration subset corresponds to a pre-trained sub-model. Based on each calibration subset and the corresponding trained preset sub-model, obtain the estimation interval corresponding to each calibration subset; Based on the estimation interval corresponding to each of the correction subsets, an integrated estimation interval is determined; Based on the integrated estimation interval and the preset residual algorithm, the upper quantile residual set and the lower quantile residual set are calculated and obtained respectively.

[0010] In an optional implementation, obtaining the lower bound and upper bound of the energy consumption baseline corresponding to the target public building based on the calibration set, the test set, and the trained preset sub-model includes: The test set is divided into multiple test subsets, and each test subset corresponds to a pre-trained sub-model; Based on each test subset and the corresponding trained preset sub-model, obtain the test estimation interval corresponding to each test subset; Determine the integration test estimation interval based on the test estimation interval corresponding to each of the test subsets; By using the integrated test estimation interval, the upper quantile residual set and the lower quantile residual set are updated respectively to obtain the lower bound of the energy consumption baseline and the upper bound of the energy consumption baseline corresponding to the target public building.

[0011] In an optional implementation, before calculating the lower boundary of indoor thermal resistance corresponding to the lower boundary of the energy consumption baseline and the upper boundary of indoor thermal resistance corresponding to the upper boundary of the energy consumption baseline based on the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, and a preset thermal parameter algorithm, the method further includes: Determine the confidence levels of the lower bound and the upper bound of the energy consumption baseline; The step of calculating and obtaining the lower boundary of indoor thermal resistance corresponding to the lower boundary of the energy consumption baseline and the upper boundary of indoor thermal resistance corresponding to the upper boundary of the energy consumption baseline based on the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, and a preset thermal parameter algorithm includes: Based on the confidence level, the lower bound of the energy consumption baseline, and the preset thermal parameter algorithm, the lower bound of the indoor thermal resistance corresponding to the lower bound of the energy consumption baseline is calculated and obtained. Based on the confidence level, the upper limit of the energy consumption baseline, and the preset thermal parameter algorithm, the upper limit of the indoor thermal resistance corresponding to the upper limit of the energy consumption baseline is calculated and obtained.

[0012] In an optional implementation, the step of calculating and obtaining the credible demand response potential range corresponding to the target public building based on the preset temperature deviation range, the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, the lower boundary of the indoor thermal resistance, the upper boundary of the indoor thermal resistance, and a preset evaluation algorithm includes: Obtain the initial reference temperature corresponding to the target public building, and determine the response target temperature based on the preset temperature deviation range and the initial reference temperature; Based on the target response temperature, the lower limit of indoor thermal resistance, and the upper limit of indoor thermal resistance, obtain the updated lower limit of energy consumption baseline and the updated upper limit of energy consumption baseline corresponding to the target response temperature. The confidence levels of the updated lower limit of energy consumption baseline and the updated upper limit of energy consumption baseline are the same as the confidence levels of the lower limit of energy consumption baseline and the upper limit of energy consumption baseline. Based on the lower bound of the energy consumption baseline, the upper bound of the energy consumption baseline, the lower bound of the updated energy consumption baseline, the upper bound of the updated energy consumption baseline, and the preset evaluation algorithm, the range of credible demand response potential corresponding to the target public building is calculated and obtained.

[0013] Secondly, the present invention provides a reliable demand response potential assessment device, comprising: The acquisition module is used to acquire multiple datasets corresponding to the target public building. The multiple datasets include: a training set, a calibration set, and a test set. The training set and the test set each include multiple sets of energy consumption factor data and historical energy consumption data corresponding to each set of energy consumption factor data. The calibration set is the relative complement of the training set. The training module is used to train multiple preset sub-models using the training set and obtain multiple trained preset sub-models. The energy consumption module is used to obtain the lower bound and upper bound of the energy consumption baseline corresponding to the target public building based on the calibration set, the test set, and the trained preset sub-model. The calculation module is used to calculate and obtain the lower limit of indoor thermal resistance corresponding to the lower limit of the energy consumption baseline and the upper limit of indoor thermal resistance corresponding to the upper limit of the energy consumption baseline based on the lower limit of the energy consumption baseline, the upper limit of the energy consumption baseline and the preset thermal parameter algorithm, respectively. The evaluation module is used to calculate and obtain the credible demand response potential range corresponding to the target public building based on the preset temperature deviation range, the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, the lower boundary of the indoor thermal resistance, the upper boundary of the indoor thermal resistance, and the preset evaluation algorithm.

[0014] Thirdly, the present invention provides an electronic device, comprising: a processor, a storage medium, and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of any of the methods described in the foregoing embodiments.

[0015] Fourthly, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the method as described in any of the foregoing embodiments.

[0016] The beneficial effects of this application are: The reliable demand response potential assessment method provided in this application includes: acquiring multiple datasets corresponding to a target public building, wherein the multiple datasets include: a training set, a calibration set, and a test set, wherein the training set and the test set each include multiple sets of energy consumption factor data and historical energy consumption data corresponding to each set of energy consumption factor data, and the calibration set is the relative complement of the training set; training multiple preset sub-models through the training set to obtain multiple trained preset sub-models; obtaining the lower bound and upper bound of the energy consumption baseline corresponding to the target public building based on the calibration set, the test set, and the trained preset sub-models; calculating the lower bound of the indoor thermal resistance corresponding to the lower bound of the energy consumption baseline and the upper bound of the indoor thermal resistance corresponding to the upper bound of the energy consumption baseline based on the lower bound of the energy consumption baseline, the upper bound of the energy consumption baseline, and a preset thermal parameter algorithm; and calculating the reliable demand response potential range corresponding to the target public building based on a preset temperature deviation range, the lower bound of the energy consumption baseline, the upper bound of the energy consumption baseline, the lower bound of the indoor thermal resistance, the upper bound of the indoor thermal resistance, and a preset assessment algorithm. In this embodiment, multiple preset sub-models are trained using a training set that includes multiple sets of energy consumption factor data and historical energy consumption data corresponding to each set of energy consumption factor data. Based on the generated historical residuals, the upper and lower bounds of the energy consumption baseline corresponding to the target public building are obtained through the trained preset sub-models. These are then combined with a preset thermal parameter algorithm to calculate the upper and lower bounds of the indoor thermal resistance corresponding to the upper and lower bounds of the energy consumption baseline. Finally, based on the upper and lower bounds of the energy consumption baseline, the upper and lower bounds of the indoor thermal resistance, temperature-related comfort constraints, and a preset evaluation algorithm, the credible demand response potential of the target public building is obtained in the form of upper and lower bound ranges. This reduces the deviation between the demand response potential evaluation results based on the gray box model and the actual response, improves the benefits of user participation in demand response, and enhances the reliability of power system dispatch. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A schematic diagram of a reliable demand response potential assessment method provided in an embodiment of this application; Figure 2 A schematic diagram of a reliable demand response potential assessment method provided in another embodiment of this application; Figure 3 A schematic flowchart of a reliable demand response potential assessment method provided in another embodiment of this application; Figure 4 A schematic diagram of a reliable demand response potential assessment method provided in another embodiment of this application; Figure 5 A schematic flowchart of a reliable demand response potential assessment method provided in another embodiment of this application; Figure 6 A schematic diagram of the energy consumption probability baseline range under different confidence levels provided in an embodiment of this application; Figure 7 A schematic diagram of a reliable demand response potential assessment device provided in this application embodiment; Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0020] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0021] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0022] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0023] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0024] Current demand response potential assessment schemes are increasingly replacing white-box and black-box models with gray-box models. White-box models, built upon physical equations, offer strong interpretability but rely on precise parameters. In practical applications, it's difficult to continuously acquire a large number of precise parameters, thus limiting the feasibility of white-box-based demand response potential assessment schemes. Black-box models, trained on large datasets, are highly adaptable but lack interpretability. Therefore, if black-box-based assessments show deviations in certain scenarios, it's difficult to pinpoint the cause and effectively correct the errors. Gray-box models, on the other hand, combine the advantages of both black-box and white-box models, offering a degree of interpretability and adaptability.

[0025] However, while gray-box models achieve a certain balance between interpretability and accuracy, existing gray-box models mostly employ deterministic outputs, i.e., outputting a single potential value, making it difficult to quantify the impact of multiple uncertainties such as weather changes, population movement, and temporal evolution on demand response potential. This leads to a significant deviation between the demand response potential assessment results under current demand response potential assessment schemes and the actual response, affecting the benefits for users participating in response activities and the reliability of system scheduling.

[0026] To address the aforementioned issues, the main objective of this application is to propose a reliable demand response potential assessment method. This method aims to reduce the discrepancy between the demand response potential assessment results based on the gray box model and the actual response, thereby improving the benefits of user participation in demand response and the reliability of power system dispatch.

[0027] Figure 1 This is a schematic flowchart of a reliable demand response potential assessment method provided in one embodiment of this application. This method can be applied, for example, to computers, servers, or other computing devices in the control rooms or dispatch rooms of power plants or other power supply units, but is not limited thereto. Figure 1 As shown, the above method includes: S101. Obtain multiple datasets corresponding to the target public building. The multiple datasets include: training set, calibration set, and test set. The training set and the test set each include multiple sets of energy consumption factor data and historical energy consumption data corresponding to each set of energy consumption factor data. The calibration set is the relative complement of the training set.

[0028] For example, the aforementioned target public building may refer to office buildings, but is not limited to office buildings. It is understood that, under normal circumstances, the main electricity consumption scenario of the target public building is air conditioning electricity consumption. Therefore, the embodiments of this application may be based on the adjustment of air conditioning electricity consumption of the target public building to achieve demand response, but are not specifically limited to this.

[0029] The aforementioned energy consumption factor data may include various factors that could potentially affect the electricity consumption of the target public building, such as environmental factors like seasonal and climatic factors, and personnel factors like the number of people and their needs. However, the specific types and quantities of energy consumption factor data are not limited to the examples mentioned above. The historical energy consumption data corresponding to the aforementioned energy consumption factor data may refer to historical energy consumption data corresponding to the time period in which the aforementioned energy consumption factor data is located, such as 1000kWh, 2000kWh, 3000kWh, etc., but is not limited to this.

[0030] The aforementioned training and test sets each include multiple sets of energy consumption factor data and historical energy consumption data corresponding to each set of the aforementioned energy consumption factor data. For example, the aforementioned training and test sets may include multiple sets of energy consumption factor data from different periods and historical energy consumption data corresponding to each set of the aforementioned energy consumption factor data. For example, the training set may consist of multiple sets of energy consumption factor data from January 1, 2025 to April 30, 2025 and historical energy consumption data corresponding to each set of the aforementioned energy consumption factor data, while the test set may consist of multiple sets of energy consumption factor data from May 1, 2025 to June 1, 2025 and historical energy consumption data corresponding to each set of the aforementioned energy consumption factor data.

[0031] It is understandable that for multiple datasets corresponding to the target public building acquired in the same instance, the number of sets of energy consumption factor data and the corresponding historical energy consumption data in the training set and test set can be different. For example, the training set could include 300 sets of energy consumption factor data and the corresponding historical energy consumption data for each set of the aforementioned energy consumption factor data, while the test set could include 160 sets of energy consumption factor data and the corresponding historical energy consumption data for each set of the aforementioned energy consumption factor data. However, the specific types and quantities of data included in the energy consumption factor data in the training set and test set should be the same. For example, the energy consumption factor data in both the training set and test set could refer to seasonal factors, climate factors, and the number of people, but this is not a limitation.

[0032] S102. Train multiple preset sub-models using the above training set to obtain multiple trained preset sub-models.

[0033] For example, the above-mentioned training set is used to train multiple preset sub-models to obtain multiple trained preset sub-models. This could mean dividing the training set into equal parts equal to the number of preset sub-models, with each part used to train one corresponding preset sub-model. For instance, assuming there are four preset sub-models and the training set includes 300 sets of energy consumption factor data and historical energy consumption data corresponding to each set, the training set could be divided into four parts. Each part would include 75 sets of energy consumption factor data and historical energy consumption data corresponding to each set, and each part would be used to train one corresponding preset sub-model. Of course, the specific number of preset sub-models, the amount of energy consumption factor data in the training set, and the number of historical energy consumption data sets corresponding to each set of energy consumption factor data can be adjusted and determined according to the actual situation, and are not limited to the examples above.

[0034] S103. Based on the above calibration set, the above test set, and the above trained preset sub-model, obtain the lower bound and upper bound of the energy consumption baseline corresponding to the above target public building.

[0035] For example, the above-mentioned acquisition of the lower and upper bounds of the energy consumption baseline corresponding to the target public building based on the calibration set, the test set, and the trained preset sub-model can, for example, refer to inputting the data from the calibration set and the test set as input values ​​into the trained preset sub-model. The calibration set input into the trained preset sub-model can, for example, combine a preset residual algorithm to obtain the lower and upper bounds of the residual set. The test set input into the trained preset sub-model can, for example, obtain preliminary estimates of the lower and upper bounds of the energy consumption baseline.

[0036] By inputting the above-mentioned calibration set into the above-mentioned trained preset sub-model to obtain the lower bound and upper bound of the residual set, and by combining the above-mentioned test set into the above-mentioned trained preset sub-model to obtain the preliminary estimates of the lower bound and upper bound of the energy consumption baseline, the lower bound and upper bound of the energy consumption baseline corresponding to the above-mentioned target public building can be obtained.

[0037] Of course, the above is only one possible example. In practice, how to obtain the lower and upper bounds of the energy consumption baseline corresponding to the target public building based on the above calibration set, the above test set, and the above trained preset sub-model is not limited to the content of the above example. Even if the lower and upper bounds of the energy consumption baseline corresponding to the target public building are obtained by the method in the above example, the specific content and form of the above preset residual algorithm, the lower and upper bounds of the residual set obtained by inputting the calibration set into the above trained preset sub-model, and the specific calculation method for comprehensively calculating the preliminary estimates of the lower and upper bounds of the energy consumption baseline obtained by inputting the test set into the above trained preset sub-model can all be adjusted and determined according to actual needs, and are not limited here.

[0038] S104. Based on the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, and the preset thermal parameter algorithm, calculate and obtain the lower boundary of the indoor thermal resistance corresponding to the lower boundary of the energy consumption baseline and the upper boundary of the indoor thermal resistance corresponding to the upper boundary of the energy consumption baseline.

[0039] For example, the above calculation to obtain the lower boundary of indoor thermal resistance corresponding to the lower boundary of the energy consumption baseline and the upper boundary of indoor thermal resistance corresponding to the upper boundary of the energy consumption baseline can be used, for example, to determine the amount of electricity adjustment corresponding to the air conditioning temperature adjustment in the case of demand response based on the electricity adjustment of the air conditioning of the target public building. It is understood that for the electricity adjustment of other electrical appliances in the target public building, other related parameters can also be calculated, and it is not limited to calculating and obtaining the lower boundary of indoor thermal resistance corresponding to the lower boundary of the energy consumption baseline and the upper boundary of indoor thermal resistance corresponding to the upper boundary of the energy consumption baseline.

[0040] S105. Based on the preset temperature deviation range, the lower limit of the energy consumption baseline, the upper limit of the energy consumption baseline, the lower limit of the indoor thermal resistance, the upper limit of the indoor thermal resistance, and the preset evaluation algorithm, calculate and obtain the credible demand response potential range corresponding to the target public building.

[0041] For example, the above calculation of the credible demand response potential range corresponding to the target public building based on the preset temperature deviation range, the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, the lower boundary of the indoor thermal resistance, the upper boundary of the indoor thermal resistance, and the preset evaluation algorithm can be similar to the example above, referring to determining the credible demand response potential range corresponding to the air conditioning temperature adjustment in the public building based on the demand response achieved by adjusting the electricity consumption of the air conditioning in the target public building. If the credible demand response potential range corresponding to the electricity consumption adjustment of other electrical appliances in the target public building is also considered, then in addition to the preset temperature deviation range, the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, the lower boundary of the indoor thermal resistance, and the upper boundary of the indoor thermal resistance, other relevant parameters and different preset evaluation algorithms can also be used to calculate the credible demand response potential range corresponding to the target public building, but the specifics are not limited here.

[0042] The reliable demand response potential assessment method provided in this application includes: acquiring multiple datasets corresponding to a target public building, wherein the multiple datasets include: a training set, a calibration set, and a test set. The training set and the test set each include multiple sets of energy consumption factor data and historical energy consumption data corresponding to each set of energy consumption factor data. The calibration set is the relative complement of the training set. Multiple preset sub-models are trained using the training set to obtain multiple trained preset sub-models. Based on the calibration set, the test set, and the trained preset sub-models, the lower bound and upper bound of the energy consumption baseline corresponding to the target public building are obtained. Based on the lower bound, the upper bound, and a preset thermal parameter algorithm, the lower bound of the indoor thermal resistance corresponding to the lower bound and the upper bound of the indoor thermal resistance corresponding to the upper bound are calculated. Based on a preset temperature deviation range, the lower bound, the upper bound, the lower bound, the upper bound, and a preset evaluation algorithm, the reliable demand response potential range corresponding to the target public building is calculated. In this embodiment, multiple preset sub-models are trained using a training set that includes multiple sets of energy consumption factor data and historical energy consumption data corresponding to each set of the aforementioned energy consumption factor data. Based on the generated historical residuals, the upper and lower bounds of the energy consumption baseline corresponding to the target public building are obtained through the trained preset sub-models. These are then combined with a preset thermal parameter algorithm to calculate the upper and lower bounds of the indoor thermal resistance corresponding to the upper and lower bounds of the energy consumption baseline. Finally, based on the upper and lower bounds of the energy consumption baseline, the upper and lower bounds of the indoor thermal resistance, temperature-related comfort constraints, and a preset evaluation algorithm, the credible demand response potential of the target public building is obtained in the form of upper and lower bound ranges. This reduces the deviation between the demand response potential evaluation results based on the gray box model and the actual response, improves the benefits of user participation in demand response, and enhances the reliability of power system dispatch.

[0043] Optionally, in the above Figure 1 Based on the embodiments, the aforementioned energy consumption factor data may include: time data, weather data, and personnel flow data. Before obtaining multiple datasets corresponding to the target public building, the method may further include: Obtain multiple sets of the aforementioned energy consumption factor data corresponding to the target public building, and the aforementioned historical energy consumption data corresponding to each set of the aforementioned energy consumption factor data.

[0044] The aforementioned acquisition of multiple datasets corresponding to the target public building may include: Based on the time data in each set of energy consumption factor data, construct the training set and the calibration set, which include time-series features.

[0045] For example, the above-mentioned training set and correction set, which include time-series features, are constructed based on the time data in each set of energy consumption factor data. For example, they can be used to represent the changes in the above-mentioned historical energy consumption data based on time, weather and personnel movement, and thus to represent the uncertainty brought about by time, weather and personnel movement.

[0046] Figure 2 For a schematic diagram of a reliable demand response potential assessment method provided in another embodiment of this application, please refer to... Figure 2 Optionally, in the foregoing Figure 1 Based on the embodiments, the above-mentioned method of training multiple preset sub-models using the training set to obtain multiple trained preset sub-models may include: S201. Divide the above training set into multiple training subsets, each of which corresponds to a preset sub-model.

[0047] For example, the training set is divided into multiple training subsets, and each training subset corresponds to a preset sub-model, such as... Figure 1 In this embodiment, the training set is divided into an equal number of preset sub-models. Each training set after being divided into equal parts is the aforementioned training subset. Each training subset corresponds to one of the aforementioned preset sub-models, and each training subset is used to train the corresponding preset sub-model.

[0048] S202. Using the above training subsets and the preset optimization algorithm, update the hyperparameters of the preset sub-models corresponding to each of the above training subsets until the hyperparameters of the preset sub-models reach the optimization target of the preset optimization algorithm, thereby obtaining multiple preset sub-models after training.

[0049] For example, the aforementioned preset sub-model can be a TCN (Temporal Convolutional Network) model with initial hyperparameters. Based on this, each preset sub-model can be represented as follows:

[0050] in, These represent multiple preset sub-models numbered from 0 to t. Let X represent the TCN structure of multiple preset sub-models numbered from 0 to t, where X is the subset of the TCN structure. The set of, where To input the energy consumption factor data of the preset sub-model, Y is the data including... A set of.

[0051] Based on this, the energy consumption factor data of each of the above preset sub-models are then used as input. The output features obtained after performing a convolution operation can be represented as:

[0052] Among them, the above This refers to the energy consumption factor data input to the aforementioned preset sub-model. The output features obtained after performing the convolution operation, as described above The filter, also known as the convolution kernel, represents the information attenuation rate and is a model parameter that can be adjusted according to actual conditions. The expansion rate is used to widen the receptive field of the preset sub-model, and it is a model parameter that can be adjusted according to the actual situation. represent Previous Data on energy consumption factors.

[0053] The receptive field of the aforementioned pre-defined sub-model depends on the depth of the pre-defined sub-model, and can be expressed as:

[0054] Among them, the above That is, the receptive field of the aforementioned preset sub-model, the aforementioned For use in representing The value, that is ,in The expansion rate is the preset sub-model value, while the model parameters can be adjusted according to actual conditions. The current layer number. This indicates that the depth of the input layer is 0. When i > 1, the kernel size can be captured. The kernel size preserves the characteristics of time-series data, rather than the entire input data, without increasing computational complexity. For the first The kernel size of the layer filter is a model parameter that can be adjusted according to the actual situation.

[0055] Figure 3 This is a schematic flowchart of an alarm notification processing method provided in another embodiment of this application. Optionally, please refer to... Figure 3 In the aforementioned Figure 2 Based on the embodiments, obtaining the lower bound and upper bound of the energy consumption baseline corresponding to the target public building according to the calibration set, the test set, and the trained preset sub-model may include: S301. Divide the above calibration set into multiple calibration subsets, each of which corresponds to a pre-trained sub-model.

[0056] Similar to the principle of dividing the training set into multiple training subsets in the above embodiments, dividing the calibration set into multiple calibration subsets can also refer to dividing the calibration set into multiple calibration subsets according to the number of preset sub-models after training, but is not limited to this.

[0057] S302. Based on each of the above-mentioned correction subsets and the corresponding trained preset sub-models, obtain the estimation interval corresponding to each of the above-mentioned correction subsets.

[0058] For example, the above-mentioned estimation intervals are obtained for each of the above-mentioned calibration subsets and the corresponding trained preset sub-models. For example, it can be based on the assumption that the time series error is stationary and mixed, which is the assumption that the energy consumption of public buildings satisfies the independent and identically distributed condition.

[0059] Based on this, the estimation interval corresponding to each of the above-mentioned correction subsets can be expressed as follows:

[0060] Among them, the above That is, the estimation interval corresponding to each of the above-mentioned correction subsets, the above-mentioned and They represent time respectively The previous recent One residual. and Let these represent the lower quantile residual set and the upper quantile residual set, respectively, to prevent the asymmetric distribution of estimation interval coverage error. and The coverage rates of the lower and upper quantile functions are controlled independently to ensure more effective coverage. (The above...) This refers to the test set mentioned above. This is a preset confidence level, such as 0.95, but it is not a limitation and can be adjusted and determined according to the actual situation. The above... The upper quantile is calculated from the integrated residual set, i.e., the 1st quantile. Quantiles.

[0061] S303. Determine the integrated estimation interval based on the estimation interval corresponding to each of the above-mentioned correction subsets.

[0062] For example, determining the integrated estimation interval based on the estimation interval corresponding to each of the above-mentioned correction subsets may refer to taking the average interval of the estimation intervals corresponding to each of the above-mentioned correction subsets, but is not limited thereto.

[0063] S304. Based on the above integrated estimation interval and preset residual algorithm, calculate and obtain the upper quantile residual set and the lower quantile residual set respectively.

[0064] Based on the above, the aforementioned set of upper quantile residuals and lower quantile residual set For example, the residuals can be calculated using the following formula, which can be, for example, the aforementioned preset residual algorithm:

[0065]

[0066] The above and Let represent the lower quantile and upper quantile of the residual set, respectively, both representing time. The previous recent Each residual, the above and The model corresponds to the input values. lower quantile and upper quantile The initial predicted value above, the lower quantile above and upper quantile All are based on the aforementioned preset confidence level values. Confirmed. The above. This is the correction set mentioned above, which is also the relative complement of the training set mentioned above.

[0067] Figure 4 This is a schematic flowchart of an alarm notification processing method provided in another embodiment of this application. Optionally, as shown... Figure 4 As shown above, in the above Figure 3 Based on the embodiments, obtaining the lower bound and upper bound of the energy consumption baseline corresponding to the target public building according to the calibration set, the test set, and the trained preset sub-model may include: S401. Divide the above test set into multiple test subsets, each of which corresponds to a pre-trained sub-model.

[0068] Similar to the principle of dividing the training set into multiple training subsets in the above embodiments, dividing the test set into multiple test subsets can also refer to dividing the test set into multiple test subsets according to the number of preset sub-models after training, but is not limited to this.

[0069] S402. Based on each of the above test subsets and the corresponding trained preset sub-models, obtain the test estimation interval for each of the above test subsets.

[0070] S403. Determine the integration test estimation interval based on the test estimation interval corresponding to each of the above test subsets.

[0071] For example, the integrated test estimation interval is determined based on the test estimation interval corresponding to each of the above test subsets. For example, it may refer to taking the average interval of the test estimation intervals corresponding to each of the above test subsets, but it is not limited to this.

[0072] The aforementioned integration test estimation interval can be expressed as, for example, as: .

[0073] S404. Through the above integrated test estimation interval, update the above upper quantile residual set and the above lower quantile residual set respectively to obtain the above lower bound and the above upper bound of the energy consumption baseline corresponding to the above target public building.

[0074] For example, the above-mentioned upper quantile residual set and lower quantile residual set are updated respectively through the above-mentioned integration test estimation interval. For example, the above-mentioned integration test estimation interval can be used to update the upper quantile residual set and lower quantile residual set respectively. The latest One estimated point Used to update the above Figure 3 upper quantile residual set in the embodiment and lower quantile residual set This leads to the aforementioned lower energy consumption baseline (BLB) and upper energy consumption baseline (BUB) corresponding to the target public building.

[0075] In addition, in the aforementioned Figure 1 Based on the embodiments, before calculating and obtaining the lower boundary of indoor thermal resistance corresponding to the lower boundary of the energy consumption baseline and the upper boundary of indoor thermal resistance corresponding to the upper boundary of the energy consumption baseline according to the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, and the preset thermal parameter algorithm, the method may further include: Determine the confidence levels of the lower bound and the upper bound of the aforementioned energy consumption baseline.

[0076] For example, the confidence levels of the lower bound and the upper bound of the energy consumption baseline can be, for example, the preset confidence level values. The confidence levels of the lower bound of the energy consumption baseline and the upper bound of the aforementioned energy consumption baseline. For example, it can be 0.95, but it is not limited to this.

[0077] Furthermore, the above-mentioned calculation of the lower boundary of the energy consumption baseline and the upper boundary of the energy consumption baseline, based on the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, and the preset thermal parameter algorithm, respectively, includes: Based on the aforementioned confidence level, the aforementioned lower bound of the energy consumption baseline, and the aforementioned preset thermal parameter algorithm, the aforementioned lower bound of the indoor thermal resistance corresponding to the aforementioned lower bound of the energy consumption baseline is calculated and obtained.

[0078] Based on the aforementioned confidence level, the aforementioned upper limit of the energy consumption baseline, and the aforementioned preset thermal parameter algorithm, the aforementioned upper limit of the indoor thermal resistance corresponding to the aforementioned upper limit of the energy consumption baseline is calculated and obtained.

[0079] For example, since the embodiments of this application are based on demand response through the regulation of air conditioning power consumption in the target public building, the first-order equivalent thermal parameter model of the air conditioning in the target public building can be determined as follows:

[0080] Among them, the above For the indoor heat capacity of the target public building, the above For the indoor thermal resistance of the target public building, the above and Let represent the indoor temperature and outdoor temperature of the target public building at time t, respectively. Let t represent the cooling capacity (or heating capacity) of the air conditioning system in the target public building at time t. The above data can be obtained by consulting the nameplate and instruction manual of the air conditioning system in the target public building, or by calculating or directly measuring the parameters in the nameplate and instruction manual in conjunction with the internal parameters of the target public building. T represents a time set including multiple different times t.

[0081] Furthermore, based on the above, the cooling capacity of the air conditioning system in the aforementioned target public building can be converted into power consumption, for example, calculated using the following formula:

[0082] Among them, the above That is, the cooling capacity of the target public building's air conditioning at time t. The corresponding power consumption, as mentioned above The heat transfer efficiency of the air conditioning system in a target public building can be obtained by consulting the nameplate and instruction manual of the air conditioning unit, or by calculating or directly measuring the efficiency based on the parameters in the nameplate and instruction manual combined with the internal parameters of the target public building. This is understandable. Under fixed conditions, the heat transfer efficiency of the air conditioning system in the aforementioned target public building The larger the value, the lower the air conditioning energy consumption of the target public building.

[0083] Based on the above, when the air conditioning in the target public building is in a stable state, Reaching the initial set temperature Then we have:

[0084] Among them, the above Confidence level The lower or upper limit of the indoor thermal resistance, where confidence level 1- The lower bound of the energy consumption baseline replaces the above. At that time, the above The lower bound of the indoor thermal resistance corresponding to the lower bound of the energy consumption baseline is calculated using a confidence level of 1- The lower limit of the energy consumption baseline replaces the above. At that time, the above The upper limit of the indoor thermal resistance is the upper limit of the energy consumption baseline.

[0085] Figure 5 Please refer to the schematic diagram of the alarm notification processing method provided in another embodiment of this application. Figure 5 Optionally, based on the above embodiments, the calculation of the credible demand response potential range corresponding to the target public building according to the preset temperature deviation range, the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, the lower boundary of the indoor thermal resistance, the upper boundary of the indoor thermal resistance, and the preset evaluation algorithm may include: S501. Obtain the initial reference temperature corresponding to the target public building, and determine the response target temperature based on the preset temperature deviation range and the initial reference temperature.

[0086] For example, the initial reference temperature corresponding to the aforementioned target public building can be, for instance, the aforementioned initial set temperature. To ensure the comfort of users within the target public building when it participates in demand response, a preset temperature deviation range can also be set. When a target public building participates in demand response, the initial set temperature of the target public building's air conditioning, also known as the initial reference temperature, should be used. Adjusted to respond to target temperature The above That is Adjusted to Temperature change value.

[0087] S502. Based on the above-mentioned target response temperature, the above-mentioned lower boundary of indoor thermal resistance and the above-mentioned upper boundary of indoor thermal resistance, obtain the updated energy consumption baseline lower boundary and the updated energy consumption baseline upper boundary corresponding to the target response temperature. The confidence levels of the above-mentioned updated energy consumption baseline lower boundary and the above-mentioned updated energy consumption baseline upper boundary are the same as the above-mentioned confidence levels of the above-mentioned lower boundary of energy consumption and the above-mentioned upper boundary of energy consumption.

[0088] For example, the above-mentioned lower bound and upper bound of the updated energy consumption baseline corresponding to the target temperature can be obtained based on the target temperature, the lower bound of the indoor thermal resistance, and the upper bound of the indoor thermal resistance, for example, through the following formula:

[0089]

[0090] Among them, the above and These are the lower bound and upper bound of the updated energy consumption baseline corresponding to the target temperature, respectively.

[0091] S503. Based on the aforementioned lower bound of the energy consumption baseline, the aforementioned upper bound of the energy consumption baseline, the aforementioned lower bound of the updated energy consumption baseline, the aforementioned upper bound of the updated energy consumption baseline, and the aforementioned preset evaluation algorithm, calculate and obtain the credible demand response potential range corresponding to the aforementioned target public building.

[0092] For example, the calculation to obtain the credible demand response potential range corresponding to the target public building can be achieved by the following formula, which can be, for example, the aforementioned preset evaluation algorithm:

[0093]

[0094]

[0095]

[0096] Among them, the upper bound of the aforementioned credible demand response potential and the lower bound of the reliable demand response potential Scope of composition This refers to the credible demand response potential range corresponding to the aforementioned target public building. It can be understood that the confidence level of this credible demand response potential range is [insert confidence level here]. .

[0097] To further explain the above Figures 1-5Example: Suppose a reliable demand response potential assessment is performed on an office building in a certain location using the reliable demand response potential assessment method described in the above embodiments of this application. When obtaining multiple datasets corresponding to the office building, both the training set and the test set include multiple sets of time data, weather data, and personnel flow data, as well as air conditioning energy consumption data corresponding to each set of time data, weather data, and personnel flow data. The time span of this data is from May 2021 to September 2021, with a time granularity of 15 minutes. Among them, the data from May 2021 to August 2021 is the training set, September 1, 2021 to September 15, 2021 is reserved as the validation set, and the data from September 16, 2021 to September 30, 2021 is the test set.

[0098] Figure 6 This is a schematic diagram of the energy consumption probability baseline intervals at different confidence levels provided in an embodiment of this application. The energy consumption probability baseline intervals at different confidence levels in the figure are obtained through the methods described above. Figure 3 and Figure 4 The steps in the embodiment, which yielded the lower bound of the energy consumption baseline at different confidence levels and the upper bound of the energy consumption baseline as described above, are used for reference. Figure 6 This indicates that the estimated interval for non-working hours is shorter than that for working hours, suggesting a significant reduction in uncertainty during non-working hours. Furthermore, the width of the energy consumption interval for working days is smaller than that for non-working days. Wider intervals are used at lower confidence levels to ensure effective coverage.

[0099] exist Figure 6 Based on the data, it can be seen that due to the influence of employee activities within the office building, the CDRP (Credible demand response potential) on holidays is lower than that on weekdays. On weekdays, the sudden increase in CDRP is due to the higher adjustment potential caused by employees failing to promptly reset the air conditioning temperature in response to a drop in ambient temperature. In contrast, although the CDRP also increases on holidays, its absolute value remains smaller. At different confidence levels, The average increase was 57.933 kW, which is 2.61 times the 22.190 kW during holidays. On average, the increase was 76.867 kW, 2.32 times the 33.121 kW increase during holidays. Overall, for public buildings, the decreases and increases in CDRP during holidays were similar. However, on weekdays, the increases... It is 1.93 times the reduction in CDRP, and the increase is... That's 1.82 times the reduction in CDRP.

[0100] Figure 6 The specific data can be found in Tables 1 and 2 below:

[0101] Table 1. Evaluation results of CDRP downsizing at different confidence intervals

[0102] Table 2. Evaluation results of CDRP adjustment at different confidence intervals Taking the CDRP results at the 95% confidence level for September 19 and 20, 2021 as examples, during the holiday period, the average daily downward adjustment range of CDRP was as follows: [12.198, 20.074] when the temperature decreased by 1℃, [30.200, 51.282] when it decreased by 2℃, and [36.594, 60.223] when it decreased by 3℃. The average upward adjustment range of CDRP was as follows: [10.770, 19.477] when the temperature increased by 1℃, [21.541, 38.953] when it increased by 2℃, and [32.311, 58.430] when it increased by 3℃. On weekdays, the average daily downward adjustment range for CDRP is as follows: [12.413, 21.869] when the temperature decreases by 1℃, [24.827, 43.738] when it decreases by 2℃, and [37.240, 65.607] when it decreases by 3℃. The average upward adjustment range for CDRP is as follows: [23.844, 37.637] when the temperature increases by 1℃, [47.688, 75.274] when it increases by 2℃, and [71.532, 112.911] when it increases by 3℃.

[0103] It should be noted that the upward adjustment of the CDRP range refers to the CDRP range in which the energy consumption of office building air conditioning decreases as the set temperature increases. The downward adjustment of the CDRP range refers to the range in which the energy consumption of office building air conditioning increases as the set temperature decreases.

[0104] Figure 7 This is a schematic diagram of a reliable demand response potential assessment device provided in an embodiment of this application. This reliable demand response potential assessment device can execute the aforementioned reliable demand response potential assessment method. The device can be integrated into computers, servers, or other computing devices in the control room or dispatch room of the power plant or other power supply unit. Figure 7 As shown, the device may include: The acquisition module 710 is used to acquire multiple datasets corresponding to the target public building. The multiple datasets include a training set, a calibration set, and a test set. The training set and the test set each include multiple sets of energy consumption factor data and historical energy consumption data corresponding to each set of energy consumption factor data. The calibration set is the relative complement of the training set.

[0105] Training module 720 is used to train multiple preset sub-models using the above training set and obtain multiple trained preset sub-models.

[0106] The energy consumption module 730 is used to obtain the lower bound and upper bound of the energy consumption baseline corresponding to the target public building based on the above-mentioned calibration set, the above-mentioned test set and the above-mentioned trained preset sub-model.

[0107] The calculation module 740 is used to calculate and obtain the lower limit of indoor thermal resistance corresponding to the lower limit of the energy consumption baseline and the upper limit of indoor thermal resistance corresponding to the upper limit of the energy consumption baseline, respectively, based on the lower limit of the energy consumption baseline, the upper limit of the energy consumption baseline and the preset thermal parameter algorithm.

[0108] The evaluation module 750 is used to calculate and obtain the credible demand response potential range corresponding to the target public building based on the preset temperature deviation range, the lower limit of the energy consumption baseline, the upper limit of the energy consumption baseline, the lower limit of the indoor thermal resistance, the upper limit of the indoor thermal resistance, and the preset evaluation algorithm.

[0109] The reliable demand response potential assessment method provided in this application includes: acquiring multiple datasets corresponding to a target public building, wherein the multiple datasets include: a training set, a calibration set, and a test set. The training set and the test set each include multiple sets of energy consumption factor data and historical energy consumption data corresponding to each set of energy consumption factor data. The calibration set is the relative complement of the training set. Multiple preset sub-models are trained using the training set to obtain multiple trained preset sub-models. Based on the calibration set, the test set, and the trained preset sub-models, the lower bound and upper bound of the energy consumption baseline corresponding to the target public building are obtained. Based on the lower bound, the upper bound, and a preset thermal parameter algorithm, the lower bound of the indoor thermal resistance corresponding to the lower bound and the upper bound of the indoor thermal resistance corresponding to the upper bound are calculated. Based on a preset temperature deviation range, the lower bound, the upper bound, the lower bound, the upper bound, and a preset evaluation algorithm, the reliable demand response potential range corresponding to the target public building is calculated. In this embodiment, multiple preset sub-models are trained using a training set that includes multiple sets of energy consumption factor data and historical energy consumption data corresponding to each set of the aforementioned energy consumption factor data. Based on the generated historical residuals, the upper and lower bounds of the energy consumption baseline corresponding to the target public building are obtained through the trained preset sub-models. These are then combined with a preset thermal parameter algorithm to calculate the upper and lower bounds of the indoor thermal resistance corresponding to the upper and lower bounds of the energy consumption baseline. Finally, based on the upper and lower bounds of the energy consumption baseline, the upper and lower bounds of the indoor thermal resistance, temperature-related comfort constraints, and a preset evaluation algorithm, the credible demand response potential of the target public building is obtained in the form of upper and lower bound ranges. This reduces the deviation between the demand response potential evaluation results based on the gray box model and the actual response, improves the benefits of user participation in demand response, and enhances the reliability of power system dispatch.

[0110] Optionally, the aforementioned energy consumption factor data includes: time data, weather data, and personnel movement data.

[0111] The aforementioned acquisition module 710 can also be used to acquire multiple sets of the aforementioned energy consumption factor data corresponding to the aforementioned target public building and the aforementioned historical energy consumption data corresponding to each set of the aforementioned energy consumption factor data.

[0112] The aforementioned reliable demand response potential assessment device further includes: a construction module, used to construct the aforementioned training set and the aforementioned correction set, which include time-series features, based on the aforementioned time data in each set of the aforementioned energy consumption factor data.

[0113] Optionally, the training module 720 is specifically used to divide the training set into multiple training subsets, each of which corresponds to a preset sub-model. Using the training subsets and a preset optimization algorithm, the hyperparameters of the preset sub-model corresponding to each training subset are updated until the hyperparameters of the preset sub-model reach the optimization target of the preset optimization algorithm, thus obtaining multiple trained preset sub-models.

[0114] Optionally, the energy consumption module 730 is specifically used to divide the calibration set into multiple calibration subsets, each calibration subset corresponding to a pre-trained sub-model. Based on each calibration subset and its corresponding pre-trained sub-model, an estimation interval is obtained for each calibration subset. Based on the estimation interval for each calibration subset, an ensemble estimation interval is determined. Based on the ensemble estimation interval and a pre-defined residual algorithm, an upper quantile residual set and a lower quantile residual set are calculated and obtained.

[0115] Optionally, the energy consumption module 730 is specifically used to divide the test set into multiple test subsets, each of which corresponds to a pre-trained sub-model. Based on each test subset and its corresponding pre-trained sub-model, a test estimation interval is obtained for each test subset. An integrated test estimation interval is determined based on the test estimation interval for each test subset. Using the integrated test estimation interval, the upper quantile residual set and the lower quantile residual set are updated to obtain the lower bound and upper bound of the energy consumption baseline for the target public building.

[0116] Optionally, the above-mentioned calculation module 740 can also be used to determine the confidence levels of the lower bound of the energy consumption baseline and the upper bound of the energy consumption baseline.

[0117] The aforementioned calculation module 740 is specifically used to calculate and obtain the lower boundary of the indoor thermal resistance corresponding to the lower boundary of the energy consumption baseline based on the aforementioned confidence level, the lower boundary of the energy consumption baseline, and the aforementioned preset thermal parameter algorithm. It also calculates and obtains the upper boundary of the indoor thermal resistance corresponding to the upper boundary of the energy consumption baseline based on the aforementioned confidence level, the upper boundary of the energy consumption baseline, and the aforementioned preset thermal parameter algorithm.

[0118] Optionally, the aforementioned assessment module 750 is specifically used to obtain the initial reference temperature corresponding to the target public building, and determine the response target temperature based on the aforementioned preset temperature deviation range and the aforementioned initial reference temperature. Based on the aforementioned response target temperature, the aforementioned lower boundary of indoor thermal resistance, and the aforementioned upper boundary of indoor thermal resistance, the updated energy consumption baseline lower boundary and the updated energy consumption baseline upper boundary corresponding to the response target temperature are obtained, wherein the confidence levels of the aforementioned updated energy consumption baseline lower boundary and the aforementioned updated energy consumption baseline upper boundary are the same as the aforementioned confidence levels of the aforementioned energy consumption baseline lower boundary and the aforementioned energy consumption baseline upper boundary. Based on the aforementioned energy consumption baseline lower boundary, the aforementioned energy consumption baseline upper boundary, the aforementioned updated energy consumption baseline lower boundary, the aforementioned updated energy consumption baseline upper boundary, and the aforementioned preset assessment algorithm, the credible demand response potential range corresponding to the aforementioned target public building is calculated and obtained.

[0119] The above-described device is used to execute the method provided in the foregoing embodiments, and its implementation principle and technical effect are similar, so they will not be described again here.

[0120] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device can be a computer, server, or other device with computing power in the control room or dispatch room of the aforementioned power plant or other power supply unit. Figure 8 As shown, the device 800 includes: The processor 810, storage medium 820, and bus 830 are connected via bus 830.

[0121] The storage medium 820 stores machine-readable instructions that can be executed by the processor 810. When the electronic device is running, the processor 810 executes the machine-readable instructions to perform the trusted demand response potential assessment method.

[0122] It should be understood that, Figure 8 The structure shown is only a schematic diagram of an electronic device; the electronic device may also include components that are larger than those shown. Figure 8 The more or fewer components shown, or having the same Figure 8 The different configurations shown. Figure 8 The components shown can be implemented using hardware, software, or a combination thereof.

[0123] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the reliable demand response potential assessment method described in the above method embodiments.

[0124] Computer-readable storage media can be electronic storage devices such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Optionally, computer-readable storage media includes non-transitory computer-readable storage medium. The computer-readable storage medium has storage space for program code that performs any of the method steps described above. This program code can be read from or written to one or more computer program exhibits. The program code can be compressed, for example, in a suitable form.

[0125] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program exhibits according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0126] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0127] If the functionality is implemented as a software module and sold or used as an independent exhibit, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software exhibit. This computer software exhibit is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0128] The above description is merely a preferred embodiment of this application and does not limit the patent scope of this application. Any equivalent structural transformations made based on the inventive concept of this application and the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included within the patent protection scope of this application.

Claims

1. A method for assessing credible demand response potential, characterized in that, include: Obtain multiple datasets corresponding to the target public building, wherein the multiple datasets include: a training set, a calibration set, and a test set. The training set and the test set each include multiple sets of energy consumption factor data and historical energy consumption data corresponding to each set of energy consumption factor data. The calibration set is the relative complement of the training set. Multiple preset sub-models are trained using the training set to obtain multiple trained preset sub-models. Based on the calibration set, the test set, and the trained preset sub-model, obtain the lower bound and upper bound of the energy consumption baseline corresponding to the target public building; Based on the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, and the preset thermal parameter algorithm, the lower boundary of the indoor thermal resistance corresponding to the lower boundary of the energy consumption baseline and the upper boundary of the indoor thermal resistance corresponding to the upper boundary of the energy consumption baseline are calculated and obtained respectively. Based on the preset temperature deviation range, the lower limit of the energy consumption baseline, the upper limit of the energy consumption baseline, the lower limit of the indoor thermal resistance, the upper limit of the indoor thermal resistance, and the preset evaluation algorithm, the range of credible demand response potential corresponding to the target public building is calculated and obtained.

2. The reliable demand response potential assessment method according to claim 1, characterized in that, The energy consumption factor data includes: time data, weather data, and personnel flow data; Before obtaining multiple datasets corresponding to the target public building, the method further includes: Obtain multiple sets of energy consumption factor data corresponding to the target public building and the historical energy consumption data corresponding to each set of energy consumption factor data; The acquisition of multiple datasets corresponding to the target public building includes: Based on the time data in each set of energy consumption factor data, a training set and a calibration set including time-series features are constructed.

3. The reliable demand response potential assessment method according to claim 1, characterized in that, The step of training multiple preset sub-models using the training set to obtain multiple trained preset sub-models includes: The training set is divided into multiple training subsets, and each training subset corresponds to a preset sub-model; The hyperparameters of the preset sub-model corresponding to each training subset are updated using the training subset and the preset optimization algorithm, until the hyperparameters of the preset sub-model reach the optimization target of the preset optimization algorithm, thereby obtaining multiple trained preset sub-models.

4. The reliable demand response potential assessment method according to claim 3, characterized in that, The step of obtaining the lower bound and upper bound of the energy consumption baseline corresponding to the target public building based on the calibration set, the test set, and the trained preset sub-model includes: The calibration set is divided into multiple calibration subsets, and each calibration subset corresponds to a pre-trained sub-model. Based on each calibration subset and the corresponding trained preset sub-model, obtain the estimation interval corresponding to each calibration subset; Based on the estimation interval corresponding to each of the correction subsets, an integrated estimation interval is determined; Based on the integrated estimation interval and the preset residual algorithm, the upper quantile residual set and the lower quantile residual set are calculated and obtained respectively.

5. The reliable demand response potential assessment method according to claim 4, characterized in that, The step of obtaining the lower bound and upper bound of the energy consumption baseline corresponding to the target public building based on the calibration set, the test set, and the trained preset sub-model includes: The test set is divided into multiple test subsets, and each test subset corresponds to a pre-trained sub-model; Based on each test subset and the corresponding trained preset sub-model, obtain the test estimation interval corresponding to each test subset; Determine the integration test estimation interval based on the test estimation interval corresponding to each of the test subsets; By using the integrated test estimation interval, the upper quantile residual set and the lower quantile residual set are updated respectively to obtain the lower bound of the energy consumption baseline and the upper bound of the energy consumption baseline corresponding to the target public building.

6. The reliable demand response potential assessment method according to claim 1, characterized in that, Before calculating and obtaining the lower boundary of indoor thermal resistance corresponding to the lower boundary of the energy consumption baseline and the upper boundary of indoor thermal resistance corresponding to the upper boundary of the energy consumption baseline based on the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, and the preset thermal parameter algorithm, the method further includes: Determine the confidence levels of the lower bound and the upper bound of the energy consumption baseline; The step of calculating and obtaining the lower boundary of indoor thermal resistance corresponding to the lower boundary of the energy consumption baseline and the upper boundary of indoor thermal resistance corresponding to the upper boundary of the energy consumption baseline based on the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, and a preset thermal parameter algorithm includes: Based on the confidence level, the lower bound of the energy consumption baseline, and the preset thermal parameter algorithm, the lower bound of the indoor thermal resistance corresponding to the lower bound of the energy consumption baseline is calculated and obtained. Based on the confidence level, the upper limit of the energy consumption baseline, and the preset thermal parameter algorithm, the upper limit of the indoor thermal resistance corresponding to the upper limit of the energy consumption baseline is calculated and obtained.

7. The reliable demand response potential assessment method according to claim 6, characterized in that, The step of calculating and obtaining the credible demand response potential range corresponding to the target public building based on the preset temperature deviation range, the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, the lower boundary of the indoor thermal resistance, the upper boundary of the indoor thermal resistance, and the preset evaluation algorithm includes: Obtain the initial reference temperature corresponding to the target public building, and determine the response target temperature based on the preset temperature deviation range and the initial reference temperature; Based on the target response temperature, the lower limit of indoor thermal resistance, and the upper limit of indoor thermal resistance, obtain the updated lower limit of energy consumption baseline and the updated upper limit of energy consumption baseline corresponding to the target response temperature. The confidence levels of the updated lower limit of energy consumption baseline and the updated upper limit of energy consumption baseline are the same as the confidence levels of the lower limit of energy consumption baseline and the upper limit of energy consumption baseline. Based on the lower bound of the energy consumption baseline, the upper bound of the energy consumption baseline, the lower bound of the updated energy consumption baseline, the upper bound of the updated energy consumption baseline, and the preset evaluation algorithm, the range of credible demand response potential corresponding to the target public building is calculated and obtained.

8. A reliable demand response potential assessment device, characterized in that, include: The acquisition module is used to acquire multiple datasets corresponding to the target public building. The multiple datasets include: a training set, a calibration set, and a test set. The training set and the test set each include multiple sets of energy consumption factor data and historical energy consumption data corresponding to each set of energy consumption factor data. The calibration set is the relative complement of the training set. The training module is used to train multiple preset sub-models using the training set and obtain multiple trained preset sub-models. The energy consumption module is used to obtain the lower bound and upper bound of the energy consumption baseline corresponding to the target public building based on the calibration set, the test set, and the trained preset sub-model. The calculation module is used to calculate and obtain the lower limit of indoor thermal resistance corresponding to the lower limit of the energy consumption baseline and the upper limit of indoor thermal resistance corresponding to the upper limit of the energy consumption baseline based on the lower limit of the energy consumption baseline, the upper limit of the energy consumption baseline and the preset thermal parameter algorithm, respectively. The evaluation module is used to calculate and obtain the credible demand response potential range corresponding to the target public building based on the preset temperature deviation range, the lower boundary of the energy consumption baseline, the upper boundary of the energy consumption baseline, the lower boundary of the indoor thermal resistance, the upper boundary of the indoor thermal resistance, and the preset evaluation algorithm.

9. An electronic device, characterized in that, include: The device includes a processor, a storage medium, and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is in operation, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the method as described in any one of claims 1-7.

Images