Thermal load regulation potential evaluation method and device based on physical mechanism, medium and equipment

Abstract

The invention discloses a thermal load regulation potential evaluation method and device based on a physical mechanism, a medium and equipment, and the method comprises the steps: obtaining to-be-predicted thermal load data, and constructing to-be-predicted thermal load features based on the to-be-predicted thermal load data, the to-be-predicted thermal load features comprising thermal load time sequence features and time period features; predicting the to-be-predicted thermal load characteristics through a pre-trained XGBoost model so as to obtain thermal load prediction data; and acquiring a thermal inertia model constructed based on a thermal inertia physical mechanism, and evaluating the thermal load regulation potential based on the thermal inertia model and the thermal load prediction data by taking a preset comfortable temperature range as a constraint.

Description

Technical Field

[0001] This application relates to the field of heat load regulation technology, and in particular to a method, apparatus, medium and equipment for evaluating heat load regulation potential based on physical mechanisms. Background Technology

[0002] In the fields of energy management and building environment control, accurately assessing the potential for heat load regulation is crucial for improving energy efficiency, ensuring indoor environmental comfort, and achieving building energy conservation and emission reduction goals. With the acceleration of urbanization and the continuous growth of building energy consumption, how to scientifically and effectively regulate heat load and assess its regulation potential has become a critical issue that urgently needs to be addressed in the current building energy sector.

[0003] Currently, common heat load forecasting methods are mainly based on statistical models or simple empirical formulas. Statistical models typically rely on large amounts of historical data, building predictive models through statistical analysis of historical heat load data. However, this method has significant limitations. On the one hand, statistical models often ignore the physical mechanisms behind heat load changes, fitting only the surface of the data, resulting in poor adaptability to complex and changing real-world conditions. For example, under different meteorological conditions, changes in human activity patterns, and variations in building usage, statistical models struggle to accurately capture dynamic changes in heat load, leading to a significant decrease in prediction accuracy. On the other hand, statistical models require extensive historical data for training. For newly constructed buildings or scenarios with missing data, it is difficult to obtain sufficient data to build a reliable predictive model, thus limiting its application in practical engineering. Summary of the Invention

[0004] In view of this, embodiments of this application provide a method, apparatus, medium, and equipment for assessing the heat load regulation potential based on physical mechanisms.

[0005] According to one aspect of this application, a method for assessing heat load regulation potential based on physical mechanisms is provided, comprising: Acquire the heat load data to be predicted, and construct the heat load characteristics to be predicted based on the heat load data to be predicted, wherein the heat load characteristics to be predicted include heat load time series characteristics and time period characteristics; The heat load characteristics to be predicted are predicted using a pre-trained XGBoost model to obtain heat load prediction data. A thermal inertia model based on the physical mechanism of thermal inertia is obtained, and the potential for thermal load adjustment is evaluated based on the thermal inertia model and the predicted thermal load data, with a preset comfort temperature range as a constraint.

[0006] Optionally, the heat load time sequence characteristics include the heat load value for a preset period of time before the current time, the heat load transformation rate for the previous time, and the cumulative heat load for a preset period of time before the current time; wherein, the heat load transformation rate is (heat load value at the current time - heat load value at the previous time) / heat load value at the previous time × 100%; The time cycle characteristics include weekend characteristics, holiday characteristics, time-of-day characteristics, weekday characteristics, and quarterly characteristics.

[0007] Optionally, before predicting the heat load characteristics to be predicted using a pre-trained XGBoost model, the method further includes: Obtain thermal load training samples; The objective function is constructed based on the loss function and the regularization term; the expression for the objective function Obj is as follows: y i This represents the actual heat load value. Here, n represents the predicted heat load value, and n is the number of heat load training samples. ,Ω(f k ) represents the complexity penalty term for the k-th decision tree, T represents the number of leaf nodes in the decision tree, w represents the weight values ​​of all leaf nodes in the decision tree, γ represents the complexity control parameter, and λ represents the regularization parameter. The XGBoost model is trained based on the thermal load training samples with the goal of minimizing the objective function value.

[0008] Optionally, after obtaining the heat load prediction data, the method further includes: The heat load prediction data is grouped using a clustering method based on a preset tolerance threshold. The heat load prediction data is then compressed based on the mean of the heat load prediction data within each group, and the heat load prediction data is updated based on the compression result.

[0009] Optionally, the expression for the thermal inertia model is: C is the system heat capacity, T_in is the indoor temperature, dT_in / dt is the rate of change of indoor temperature over time, T_out is the outdoor temperature, R is the system thermal resistance, Q_heat is the heating power, and Q_loss is the heat loss power.

[0010] Optionally, constrained by a preset comfort temperature range, an assessment of the heat load regulation potential is performed based on the thermal inertia model and the predicted heat load data, including: Baseline heating capacity is determined based on updated heat load forecast data; An expression for the heating power is constructed based on the baseline heating power and the adjustment amount; wherein, the adjustment amount is a term to be solved. The expression for the heating power and the predicted outdoor temperature are substituted into the thermal inertia model. With the preset comfort temperature range as a constraint, the indoor temperature change curve is simulated to solve for the adjustment amount. The heat load adjustment potential is determined based on the adjustment amount.

[0011] Optionally, the step of simulating an indoor temperature change curve to determine the adjustment amount, constrained by a preset comfortable temperature range, includes: Using the lower limit of the preset comfortable temperature range as a constraint, the heating power is gradually reduced by adjusting the adjustment amount so that the lower limit of the indoor temperature change curve is close to the lower limit of the preset comfortable temperature range, thereby determining the potential for further adjustment of the adjustment amount. By using the upper limit of the preset comfortable temperature range as a constraint, the heating power is gradually increased by adjusting the adjustment amount so that the upper limit of the indoor temperature change curve approaches the upper limit of the preset comfortable temperature range, thereby determining the upward adjustment potential of the adjustment amount.

[0012] Optionally, determining the heat load regulation potential based on the regulation amount includes: The heat load adjustment potential is determined based on the downward and upward adjustment potential of the adjustment amount.

[0013] According to another aspect of this application, a heat load regulation potential assessment device based on physical mechanisms is provided, the device comprising: The feature construction module is used to acquire the heat load data to be predicted and construct the heat load features to be predicted based on the heat load data to be predicted, wherein the heat load features to be predicted include heat load time series features and time period features; The prediction module is used to predict the heat load characteristics to be predicted using a pre-trained XGBoost model to obtain heat load prediction data. The evaluation module is used to obtain a thermal inertia model based on the physical mechanism of thermal inertia, and to evaluate the potential for thermal load adjustment based on the thermal inertia model and the predicted heat load data, with a preset comfort temperature range as a constraint.

[0014] Optionally, the heat load time sequence characteristics include the heat load value for a preset period of time before the current time, the heat load transformation rate for the previous time, and the cumulative heat load for a preset period of time before the current time; wherein, the heat load transformation rate is (heat load value at the current time - heat load value at the previous time) / heat load value at the previous time × 100%; The time cycle characteristics include weekend characteristics, holiday characteristics, time-of-day characteristics, weekday characteristics, and quarterly characteristics.

[0015] Optionally, the apparatus further includes: a model training module, used for: Obtain thermal load training samples; The objective function is constructed based on the loss function and the regularization term; the expression for the objective function Obj is as follows: y i This represents the actual heat load value. Here, n represents the predicted heat load value, and n is the number of heat load training samples. ,Ω(f k ) represents the complexity penalty term for the k-th decision tree, T represents the number of leaf nodes in the decision tree, w represents the weight values ​​of all leaf nodes in the decision tree, γ represents the complexity control parameter, and λ represents the regularization parameter. The XGBoost model is trained based on the thermal load training samples with the goal of minimizing the objective function value.

[0016] Optionally, the prediction module is further configured to: The heat load prediction data is grouped using a clustering method based on a preset tolerance threshold. The heat load prediction data is then compressed based on the mean of the heat load prediction data within each group, and the heat load prediction data is updated based on the compression result.

[0017] Optionally, the expression for the thermal inertia model is: C is the system heat capacity, T_in is the indoor temperature, dT_in / dt is the rate of change of indoor temperature over time, T_out is the outdoor temperature, R is the system thermal resistance, Q_heat is the heating power, and Q_loss is the heat loss power.

[0018] Optionally, the evaluation module is further configured to: Baseline heating capacity is determined based on updated heat load forecast data; An expression for the heating power is constructed based on the baseline heating power and the adjustment amount; wherein, the adjustment amount is a term to be solved. The expression for the heating power and the predicted outdoor temperature are substituted into the thermal inertia model. With the preset comfort temperature range as a constraint, the indoor temperature change curve is simulated to solve for the adjustment amount. The heat load adjustment potential is determined based on the adjustment amount.

[0019] Optionally, the evaluation module is further configured to: Using the lower limit of the preset comfortable temperature range as a constraint, the heating power is gradually reduced by adjusting the adjustment amount so that the lower limit of the indoor temperature change curve is close to the lower limit of the preset comfortable temperature range, thereby determining the potential for further adjustment of the adjustment amount. By using the upper limit of the preset comfortable temperature range as a constraint, the heating power is gradually increased by adjusting the adjustment amount so that the upper limit of the indoor temperature change curve approaches the upper limit of the preset comfortable temperature range, thereby determining the upward adjustment potential of the adjustment amount.

[0020] Optionally, the evaluation module is further configured to: The heat load adjustment potential is determined based on the downward and upward adjustment potential of the adjustment amount.

[0021] According to another aspect of this application, a storage medium is provided that stores a computer program thereon, which, when executed by a processor, implements the above-described method for assessing the heat load regulation potential based on physical mechanisms.

[0022] According to another aspect of this application, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor executes the program to implement the above-described physical mechanism-based thermal load regulation potential assessment method.

[0023] By employing the above technical solutions, this application provides a method, apparatus, medium, and equipment for assessing heat load regulation potential based on physical mechanisms. This constructs a predictable heat load feature that includes both temporal and periodic characteristics, enabling a more comprehensive characterization of heat load variations. The temporal characteristics capture short-term dynamic changes in heat load, while the periodic characteristics consider long-term periodic fluctuations. The combination of these two features provides rich information for the XGBoost model, allowing it to more accurately learn the relationship between heat load and various features, thereby improving the accuracy of heat load prediction data. Furthermore, through the thermal inertia model, the change process of indoor temperature under different heat load regulation conditions can be realistically simulated, making the assessment of heat load regulation potential more consistent with actual physical conditions and accurately reflecting the trend of indoor temperature changes under different regulation strategies. By using a preset comfort temperature range as a constraint for assessment, regulation schemes that meet both energy regulation needs and ensure indoor comfort can be selected when evaluating heat load regulation potential. This avoids energy waste caused by over-regulation and prevents insufficient regulation from affecting indoor comfort, achieving a balance between energy conservation and indoor comfort.

[0024] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0025] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 A flowchart illustrating a method for assessing the heat load regulation potential based on physical mechanisms, provided in an embodiment of this application, is shown. Figure 2 A schematic diagram of a heat load regulation potential assessment device based on physical mechanisms provided in an embodiment of this application is shown. Detailed Implementation

[0026] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present application can be combined with each other.

[0027] Example 1 This embodiment provides a method for assessing the heat load regulation potential based on physical mechanisms, such as... Figure 1 As shown, the method includes: Step 101: Obtain the heat load data to be predicted, and construct the heat load characteristics to be predicted based on the heat load data to be predicted, wherein the heat load characteristics to be predicted include heat load time series characteristics and time period characteristics.

[0028] In this embodiment, firstly, the heat load data to be predicted is acquired, and then the features of the heat load to be predicted are constructed based on this data. The features constructed here cover the heat load time series features and time periodic features. The heat load time series features can reflect the continuous change of heat load over time, such as the numerical change of heat load at different times, which helps to capture the short-term dynamic changes of heat load; the time periodic features consider the periodicity of heat load over a longer time scale, such as the heat load change patterns that occur on a daily, weekly, monthly, or even seasonal basis. For example, the heat load during the winter heating season is usually high and has a certain periodic fluctuation. By extracting these features, more comprehensive and accurate information can be provided for subsequent predictions.

[0029] Optionally, the heat load time-series characteristics include the heat load value for a preset duration prior to the current moment, the heat load transformation rate for the previous moment, and the cumulative heat load for a preset duration prior to the current moment; wherein, the heat load transformation rate is (heat load value at the current moment - heat load value at the previous moment) / heat load value at the previous moment × 100%; the time period characteristics include weekend characteristics, holiday characteristics, time-of-day characteristics, weekday characteristics, and quarterly characteristics. For example: Whether it is a weekend: binary characteristic, 1 represents a weekend, 0 represents a weekday. Whether it is a holiday: binary characteristic, 1 represents a holiday, 0 represents a non-holiday. Time of day: continuous variable, representing a specific time of day (e.g., 0-23 hours). Day of the week: categorical variable, taking values ​​from 1 to 7, representing Monday to Sunday.

[0030] Step 102: Predict the heat load characteristics to be predicted using a pre-trained XGBoost model to obtain heat load prediction data.

[0031] Next, a pre-trained XGBoost model is used to predict the constructed heat load features, thereby obtaining heat load prediction data. XGBoost is an efficient ensemble learning algorithm that combines multiple weak learners (such as decision trees) to construct a strong learner, possessing the ability to handle complex nonlinear relationships. In this method, the pre-trained XGBoost model has learned the complex mapping relationship between heat load features and actual heat load, and can predict future heat load values ​​relatively accurately based on the input heat load features, providing basic data for subsequent assessment of heat load regulation potential.

[0032] Step 103: Obtain a thermal inertia model based on the physical mechanism of thermal inertia, and evaluate the potential for thermal load adjustment based on the thermal inertia model and the predicted thermal load data, with a preset comfort temperature range as a constraint.

[0033] Finally, a thermal inertia model based on the physical mechanism of thermal inertia is obtained, and the heat load adjustment potential is evaluated based on the thermal inertia model and heat load prediction data, constrained by a preset comfort temperature range. The thermal inertia model simulates the change of indoor temperature over time under different heat load adjustment conditions. The preset comfort temperature range clarifies the acceptable temperature range of the indoor environment. Using this as a constraint for evaluation ensures that heat load adjustment can not only meet the requirements of indoor comfort, but also reasonably assess the adjustable range of heat load under different conditions, i.e., the heat load adjustment potential.

[0034] By applying the technical solution of this embodiment, a predictable heat load feature including both temporal and periodic characteristics can be constructed, enabling a more comprehensive characterization of heat load variation patterns. The temporal characteristics capture short-term dynamic changes in heat load, while the periodic characteristics consider long-term periodic fluctuations. The combination of these two features provides rich information for the XGBoost model, allowing it to more accurately learn the relationship between heat load and various characteristics, thereby improving the accuracy of heat load prediction data. Furthermore, through the thermal inertia model, the indoor temperature change process under different heat load adjustment conditions can be realistically simulated, making the assessment of heat load adjustment potential more consistent with actual physical conditions and accurately reflecting the indoor temperature change trend under different adjustment strategies. By using a preset comfort temperature range as a constraint for evaluation, adjustment schemes that meet both energy adjustment needs and ensure indoor comfort can be selected when assessing heat load adjustment potential. This avoids energy waste caused by over-adjustment and prevents insufficient adjustment from affecting indoor comfort, achieving a balance between energy conservation and indoor comfort.

[0035] Example 2 In another embodiment of this application, optionally, before predicting the heat load characteristics to be predicted using a pre-trained XGBoost model, the method further includes: Obtain thermal load training samples; The objective function is constructed based on the loss function and the regularization term; the expression for the objective function Obj is as follows: y i This represents the actual heat load value. Here, n represents the predicted heat load value, and n is the number of heat load training samples. ,Ω(f k ) represents the complexity penalty term for the k-th decision tree, T represents the number of leaf nodes in the decision tree, w represents the weight values ​​of all leaf nodes in the decision tree, γ represents the complexity control parameter, and λ represents the regularization parameter. The XGBoost model is trained based on the thermal load training samples with the goal of minimizing the objective function value.

[0036] In this embodiment, the objective function consists of two parts: a loss function and a regularization term. The training process involves continuously adjusting the model parameters (i.e., the structure of the decision tree and the leaf weights) to minimize the value of this objective function, Obj(θ). The loss function measures the model's prediction accuracy. Its role is to force the model's predictions to be more accurate. As close as possible to the true observed value y iThe solution is reflected in the following: Minimizing the loss function directly ensures the high accuracy of the model's output heat load prediction. The regularization term is key to embodying "physical mechanism guidance" and ensuring the model's generalization ability. It penalizes the model's complexity, preventing it from "learning too well" (overfitting) on ​​the training data, thus learning noise in the data and the random characteristics of specific samples, resulting in poor performance on new data. Here, T is the number of leaf nodes in the decision tree. The deeper the tree and the more leaf nodes, the more complex the model. w is the weight value of all leaf nodes in the decision tree. The larger the weight value, the more "aggressive" the model's prediction. γ and λ are important hyperparameters, controlling the penalty for the number of leaf nodes and the leaf weights, respectively. γ is a complexity control parameter. The larger the γ value, the more the model tends to generate simple trees (fewer leaf nodes). λ is a regularization parameter. The larger the λ value, the greater the penalty for the leaf weights, forcing the weight values ​​to tend to be smooth, preventing the model from overreacting to individual samples. Since heat load data comes from complex physical systems, it inevitably contains noise and uncertainty. If we only pursue minimizing the loss function, the model may become excessively complex, trying to memorize every fluctuation in the training data. However, this clearly does not conform to the physical laws of thermodynamic systems, such as smoothness and inertia. By introducing a regularization term, the objective function seeks the optimal balance between "fitting accuracy" and "model simplicity," ensuring that the model learns the robust and universal physical laws behind the data, rather than surface noise. This is precisely the embodiment of "physical mechanism guidance."

[0037] Example 3 In another embodiment of this application, optionally, after obtaining the heat load prediction data, the method further includes: grouping the heat load prediction data using a clustering method based on a preset tolerance threshold, compressing the heat load prediction data based on the mean of the heat load prediction data within the group, and updating the heat load prediction data based on the compression result.

[0038] In this embodiment, after obtaining heat load prediction data through a pre-trained XGBoost model, the heat load prediction data is further processed. Specifically, a clustering method based on a preset tolerance threshold is used to group the heat load prediction data. The preset tolerance threshold is a pre-defined allowable error range. During the clustering process, data in the heat load prediction data that differ from each other within the preset tolerance threshold range are grouped into the same group. For example, if the preset tolerance threshold is set to a small value, then data with similar heat load prediction values ​​will be grouped into the same data group. After completing the data grouping, the original heat load prediction data is compressed based on the mean of the heat load prediction data within each group. That is, the average value of the data within the group is used to represent all the data within that group. In this way, the originally potentially large amount of heat load prediction data with a certain degree of similarity is simplified to a representative value for each group. Finally, the original heat load prediction data is updated based on this compression result, so that the data used in subsequent thermal inertia model evaluations based on the heat load prediction data is the compressed data, thereby improving the efficiency of subsequent evaluations.

[0039] In another embodiment of this application, optionally, the expression for the thermal inertia model is: C is the system heat capacity, T_in is the indoor temperature, dT_in / dt is the rate of change of indoor temperature over time, T_out is the outdoor temperature, R is the system thermal resistance, Q_heat is the heating power, and Q_loss is the heat loss power.

[0040] Example 4 In another embodiment of this application, optionally, the assessment of heat load adjustment potential is performed based on the thermal inertia model and the heat load prediction data, constrained by a preset comfort temperature range. This includes: determining a baseline heating power based on the updated heat load prediction data; constructing an expression for the heating power based on the baseline heating power and the adjustment amount; wherein the adjustment amount is a term to be solved; substituting the expression for the heating power and the predicted outdoor temperature into the thermal inertia model, and simulating an indoor temperature change curve under the constraint of the preset comfort temperature range, to solve for the adjustment amount; and determining the heat load adjustment potential based on the adjustment amount.

[0041] In this embodiment, firstly, the baseline heating power is determined based on the updated heat load forecast data. The updated heat load forecast data, processed as in previous embodiments, more accurately reflects future heat load changes. The baseline heating power determined based on this data serves as a reference for subsequent adjustments, representing the heating power required to maintain a certain heat load state without additional adjustments. Next, an expression for the heating power is constructed based on the baseline heating power and the adjustment amount. Here, the adjustment amount is set as the term to be solved, so that subsequent calculations can determine at what adjustment amount the goal of both satisfying the preset comfort temperature range and reasonably assessing the heat load adjustment potential can be achieved. The heating power expression links the baseline heating power and the adjustment amount, providing a mathematical basis for subsequent simulations and calculations. Then, the constructed heating power expression and the predicted outdoor temperature are substituted into the thermal inertia model. The thermal inertia model considers the physical characteristics of heat storage and release in the building envelope (i.e., heat loss power), and can simulate changes in indoor temperature under different heating power and outdoor temperature conditions. Constrained by a preset comfort temperature range, this study simulates indoor temperature changes to find a suitable adjustment amount that ensures the indoor temperature remains within the preset comfort range under given heating power and outdoor temperature conditions. Finally, the heat load adjustment potential is determined based on the calculated adjustment amount. The adjustment amount reflects the range of adjustable heating power under current conditions, and this adjustment range directly corresponds to the heat load adjustment potential—that is, the degree to which the heat load can be increased or decreased without disrupting the indoor comfort temperature range.

[0042] Example 5 In another embodiment of this application, optionally, the step of simulating an indoor temperature change curve to solve for the adjustment amount, constrained by a preset comfort temperature range, includes: using the lower limit of the preset comfort temperature range as a constraint, gradually reducing the heating power by adjusting the adjustment amount to make the lower limit of the indoor temperature change curve approach the lower limit of the preset comfort temperature range, thereby determining the downward adjustment potential of the adjustment amount; and using the upper limit of the preset comfort temperature range as a constraint, gradually increasing the heating power by adjusting the adjustment amount to make the upper limit of the indoor temperature change curve approach the upper limit of the preset comfort temperature range, thereby determining the upward adjustment potential of the adjustment amount. Wherein, determining the heat load adjustment potential based on the adjustment amount includes: determining the heat load adjustment potential based on the downward and upward adjustment potentials of the adjustment amount.

[0043] In this embodiment, on the one hand, the lower limit of the preset comfortable temperature range is used as a constraint. In actual operation, the heating power is gradually reduced by continuously adjusting the adjustment amount. As the heating power decreases, the indoor temperature will change accordingly, and the indoor temperature change curve will also change accordingly. During this process, the lower limit of the indoor temperature change curve is continuously observed, with the goal of making this lower limit as close as possible to the lower limit of the preset comfortable temperature range. When the lower limit of the indoor temperature change curve is just close to the lower limit of the preset comfortable temperature range, the change in the corresponding adjustment amount determines the potential for further reduction of the adjustment amount. This means that the maximum extent to which the heating power can be reduced without lowering the indoor temperature below the lower limit of the comfortable temperature range is the maximum value that the adjustment amount can be reduced to. On the other hand, the upper limit of the preset comfortable temperature range is used as a constraint. The heating power is gradually increased by adjusting the adjustment amount. The increase in heating power will cause the indoor temperature to rise, thereby affecting the indoor temperature change curve. The upper limit of the indoor temperature change curve is monitored, gradually bringing it closer to the upper limit of the preset comfortable temperature range. When the upper limit of the indoor temperature change curve is close to the upper limit of the preset comfortable temperature range, the change in the corresponding adjustment amount determines the potential for further increase of the adjustment amount. This refers to the maximum extent to which the heating power can be increased without causing the indoor temperature to exceed the upper limit of the comfortable temperature range; in other words, the maximum value that the adjustment amount can be increased.

[0044] In the section on determining the heat load adjustment potential based on the adjustment amount, the downward and upward adjustment potentials of the previously determined adjustment amount are comprehensively considered, clarifying the range in which the heat load can be adjusted in different directions.

[0045] By applying the technical solutions of the embodiments of this application, at least the following beneficial effects can be achieved: From the perspective of assessment accuracy, determining the baseline heating power based on the updated heat load forecast data is more precise because the updated data, after undergoing clustering and compression, more accurately reflects the actual heat load. This provides a reliable foundation for subsequently constructing the heating power expression and the entire assessment process, enabling the calculation of adjustment quantities and the assessment of heat load adjustment potential based on it to more closely reflect the actual situation and improve the accuracy of the assessment results.

[0046] In terms of physical realities, the heating power expression and predicted outdoor temperature are substituted into the thermal inertia model for simulation, fully considering the building's thermal inertia and the impact of the outdoor environment on indoor temperature. The thermal inertia model can realistically reflect the characteristics of the building envelope during heat transfer, and outdoor temperature is one of the important factors affecting the indoor thermal environment. Through this comprehensive simulation, the changes in indoor temperature under different heating power adjustments can be predicted more accurately, avoiding the shortcomings of evaluating solely from a theoretical perspective while ignoring actual physical processes, making the evaluation results more practically instructive.

[0047] By using a preset comfort temperature range as a constraint, the solution ensures indoor environmental comfort. In actual heat load regulation, it is not enough to simply pursue energy savings or maximize the heat load regulation potential; it is also essential to ensure that the indoor temperature remains within an acceptable range for users. This embodiment uses the comfort temperature range as an explicit constraint, ensuring that the indoor temperature meets the requirements throughout the solution process, thereby achieving a balance between energy regulation and indoor comfort. The heat load regulation potential evaluated in this way satisfies energy-saving requirements without adversely affecting the indoor environment.

[0048] Furthermore, as Figure 1 In terms of specific implementation, this application provides a heat load regulation potential assessment device based on physical mechanisms, such as... Figure 2 As shown, the device includes: The feature construction module is used to acquire the heat load data to be predicted and construct the heat load features to be predicted based on the heat load data to be predicted, wherein the heat load features to be predicted include heat load time series features and time period features; The prediction module is used to predict the heat load characteristics to be predicted using a pre-trained XGBoost model to obtain heat load prediction data. The evaluation module is used to obtain a thermal inertia model based on the physical mechanism of thermal inertia, and to evaluate the potential for thermal load adjustment based on the thermal inertia model and the predicted heat load data, with a preset comfort temperature range as a constraint.

[0049] Optionally, the heat load time sequence characteristics include the heat load value for a preset period of time before the current time, the heat load transformation rate for the previous time, and the cumulative heat load for a preset period of time before the current time; wherein, the heat load transformation rate is (heat load value at the current time - heat load value at the previous time) / heat load value at the previous time × 100%; The time cycle characteristics include weekend characteristics, holiday characteristics, time-of-day characteristics, weekday characteristics, and quarterly characteristics.

[0050] Optionally, the apparatus further includes: a model training module, used for: Obtain thermal load training samples; The objective function is constructed based on the loss function and the regularization term; the expression for the objective function Obj is as follows: y i This represents the actual heat load value. Here, n represents the predicted heat load value, and n is the number of heat load training samples. ,Ω(f k ) represents the complexity penalty term for the k-th decision tree, T represents the number of leaf nodes in the decision tree, w represents the weight values ​​of all leaf nodes in the decision tree, γ represents the complexity control parameter, and λ represents the regularization parameter. The XGBoost model is trained based on the thermal load training samples with the goal of minimizing the objective function value.

[0051] Optionally, the prediction module is further configured to: The heat load prediction data is grouped using a clustering method based on a preset tolerance threshold. The heat load prediction data is then compressed based on the mean of the heat load prediction data within each group, and the heat load prediction data is updated based on the compression result.

[0052] Optionally, the expression for the thermal inertia model is: C is the system heat capacity, T_in is the indoor temperature, dT_in / dt is the rate of change of indoor temperature over time, T_out is the outdoor temperature, R is the system thermal resistance, Q_heat is the heating power, and Q_loss is the heat loss power.

[0053] Optionally, the evaluation module is further configured to: Baseline heating capacity is determined based on updated heat load forecast data; An expression for the heating power is constructed based on the baseline heating power and the adjustment amount; wherein, the adjustment amount is a term to be solved. The expression for the heating power and the predicted outdoor temperature are substituted into the thermal inertia model. With the preset comfort temperature range as a constraint, the indoor temperature change curve is simulated to solve for the adjustment amount. The heat load adjustment potential is determined based on the adjustment amount.

[0054] Optionally, the evaluation module is further configured to: Using the lower limit of the preset comfortable temperature range as a constraint, the heating power is gradually reduced by adjusting the adjustment amount so that the lower limit of the indoor temperature change curve is close to the lower limit of the preset comfortable temperature range, thereby determining the potential for further adjustment of the adjustment amount. By using the upper limit of the preset comfortable temperature range as a constraint, the heating power is gradually increased by adjusting the adjustment amount so that the upper limit of the indoor temperature change curve approaches the upper limit of the preset comfortable temperature range, thereby determining the upward adjustment potential of the adjustment amount.

[0055] Optionally, the evaluation module is further configured to: The heat load adjustment potential is determined based on the downward and upward adjustment potential of the adjustment amount.

[0056] It should be noted that other corresponding descriptions of the functional units involved in the physical mechanism-based heat load regulation potential assessment device provided in this application embodiment can be found in the following references. Figure 1 The corresponding descriptions in the method will not be repeated here.

[0057] This application also provides a computer device, specifically a personal computer, server, network device, etc. The computer device includes a bus, processor, memory, and communication interface, and may also include input / output interfaces and a display device. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device stores location information. The network interface of the computer device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the steps in the various method embodiments.

[0058] Those skilled in the art will understand that the structure of the computer device described above is only a partial structure related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. A specific computer device may include more or fewer components, or combine certain components, or have different component arrangements.

[0059] In one embodiment, a computer-readable storage medium is provided, which may be non-volatile or volatile, having stored thereon a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0060] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0061] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0062] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, graphics processors, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0063] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are 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.

[0064] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for assessing the heat load regulation potential based on physical mechanisms, characterized in that, include: Acquire the heat load data to be predicted, and construct the heat load characteristics to be predicted based on the heat load data to be predicted, wherein the heat load characteristics to be predicted include heat load time series characteristics and time period characteristics; The heat load characteristics to be predicted are predicted using a pre-trained XGBoost model to obtain heat load prediction data. A thermal inertia model based on the physical mechanism of thermal inertia is obtained, and the potential for thermal load adjustment is evaluated based on the thermal inertia model and the predicted thermal load data, with a preset comfort temperature range as a constraint.

2. The method according to claim 1, characterized in that, The heat load time sequence characteristics include the heat load value for a preset period of time before the current time, the heat load transformation rate for the previous time, and the cumulative heat load for a preset period of time before the current time; wherein, the heat load transformation rate is (heat load value at the current time - heat load value at the previous time) / heat load value at the previous time × 100%; The time cycle characteristics include weekend characteristics, holiday characteristics, time-of-day characteristics, weekday characteristics, and quarterly characteristics.

3. The method according to claim 1, characterized in that, Before predicting the heat load characteristics to be predicted using a pre-trained XGBoost model, the method further includes: Obtain thermal load training samples; The objective function is constructed based on the loss function and the regularization term; the expression for the objective function Obj is as follows: y i This represents the actual heat load value. Here, n represents the predicted heat load value, and n is the number of heat load training samples. ,Ω(f k ) represents the complexity penalty term for the k-th decision tree, T represents the number of leaf nodes in the decision tree, w represents the weight values ​​of all leaf nodes in the decision tree, γ represents the complexity control parameter, and λ represents the regularization parameter. The XGBoost model is trained based on the thermal load training samples with the goal of minimizing the objective function value.

4. The method according to claim 1, characterized in that, After obtaining the heat load prediction data, the method further includes: The heat load prediction data is grouped using a clustering method based on a preset tolerance threshold. The heat load prediction data is then compressed based on the mean of the heat load prediction data within each group, and the heat load prediction data is updated based on the compression result.

5. The method according to claim 4, characterized in that, The expression for the thermal inertia model is: C is the system heat capacity, T_in is the indoor temperature, dT_in / dt is the rate of change of indoor temperature over time, T_out is the outdoor temperature, R is the system thermal resistance, Q_heat is the heating power, and Q_loss is the heat loss power.

6. The method according to claim 5, characterized in that, The assessment of heat load regulation potential, constrained by a preset comfort temperature range and based on the thermal inertia model and the predicted heat load data, includes: Baseline heating capacity is determined based on updated heat load forecast data; An expression for the heating power is constructed based on the baseline heating power and the adjustment amount; wherein, the adjustment amount is a term to be solved. The expression for the heating power and the predicted outdoor temperature are substituted into the thermal inertia model. With the preset comfort temperature range as a constraint, the indoor temperature change curve is simulated to solve for the adjustment amount. The heat load adjustment potential is determined based on the adjustment amount.

7. The method according to claim 6, characterized in that, The step of simulating indoor temperature change curves under the constraint of a preset comfortable temperature range to solve for the adjustment amount includes: Using the lower limit of the preset comfortable temperature range as a constraint, the heating power is gradually reduced by adjusting the adjustment amount so that the lower limit of the indoor temperature change curve is close to the lower limit of the preset comfortable temperature range, thereby determining the potential for further adjustment of the adjustment amount. By using the upper limit of the preset comfortable temperature range as a constraint, the heating power is gradually increased by adjusting the adjustment amount so that the upper limit of the indoor temperature change curve approaches the upper limit of the preset comfortable temperature range, thereby determining the upward adjustment potential of the adjustment amount.

8. The method according to claim 7, characterized in that, Determining the heat load regulation potential based on the regulation amount includes: The heat load adjustment potential is determined based on the downward and upward adjustment potential of the adjustment amount.

9. A device for assessing the heat load regulation potential based on physical mechanisms, characterized in that, The device includes: The feature construction module is used to acquire the heat load data to be predicted and construct the heat load features to be predicted based on the heat load data to be predicted, wherein the heat load features to be predicted include heat load time series features and time period features; The prediction module is used to predict the heat load characteristics to be predicted using a pre-trained XGBoost model to obtain heat load prediction data. The evaluation module is used to obtain a thermal inertia model based on the physical mechanism of thermal inertia, and to evaluate the potential for thermal load adjustment based on the thermal inertia model and the predicted heat load data, with a preset comfort temperature range as a constraint.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 8.

11. A computer device, comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 8.

Images