Cold and hot load online prediction method and device

By constructing a neural network model to predict indoor temperature and reversely solve the energy supply load, the accuracy and explanatory problems of predicting hot and cold loads in indoor temperature change scenarios in the prior art are solved, and short-term and ultra-short-time hot and cold loads are achieved Self-learning of load prediction and energy supply system parameters.

CN119090671APending Publication Date: 2024-12-06XIDIAN POWER RECTIFIER XIAN +1
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411112195.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-14
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The existing hot and cold load prediction models are difficult to achieve short-term and ultra-short-time accurate prediction in intermittent heating/cooling scenarios with indoor temperature changes, and lack explanatory nature.

Method used

By constructing a neural network model, the energy supply load at the current moment, outdoor environmental data and historical indoor temperature are used to predict the indoor temperature at the next moment, and the energy supply load at the current moment is reversely solved to determine the water inlet temperature of the energy supply system Average with the effluent temperature.

Benefits of technology

It realizes short- and ultra-short-time cold and hot load prediction in intermittent heating/cooling scenarios for indoor temperature changes, improves the accuracy and interpretation of the prediction, and has self-learning and self-correction functions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119090671A_ABST
    Figure CN119090671A_ABST
Patent Text Reader

Abstract

The invention discloses a cold and hot load online prediction method and device, and the method comprises the steps: obtaining a load prediction historical record comprising a preset time duration of a current moment, the load prediction historical record comprising an energy supply load of an energy supply station, and outdoor environment data and indoor temperature of a target building, and the energy supply load being a cooling load or a heating load; the load prediction historical record of the preset time duration is input into a trained indoor temperature prediction model, the indoor temperature of the next moment is predicted, and the indoor temperature prediction model takes the energy supply load, the outdoor environment parameters and the historical indoor temperature as input variables; a neural network model is constructed by taking future indoor temperature as an output variable and then is trained; and the energy supply load at the current moment is solved reversely by taking the indoor temperature at the next moment as an expected value, so that the average value of the water inlet temperature and the water outlet temperature of the energy supply system is determined. The method can be suitable for intermittent heat supply / cold supply scenes with indoor temperature changes, and short-time and ultra-short-time load prediction is provided.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of cold / hot load prediction of energy systems, and in particular to a cold / hot load online prediction method and device. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the invention recited in the claims. No admission is made that the description herein is prior art by inclusion in this section.

[0003] Central heating and cooling are often used in industrial parks, commercial complexes, and public buildings such as schools, hospitals, and libraries. This improves energy efficiency while facilitating maintenance and management. Research on cold / heat load prediction to improve the accuracy of cold / heat load prediction and achieve on-demand energy supply is one of the important means to reduce the energy consumption of heating / cooling units.

[0004] According to the different forecasting periods, the cold / heating load forecast can be divided into three types: ultra-short-term heat load forecast, short-term load forecast, and medium- and long-term load forecast. Ultra-short-term load forecast: refers to the forecast of load values ​​within tens of minutes, minutes or even seconds in the future. Short-term load forecast: refers to the load forecast from one day to one week, usually in hours. Medium- and long-term load forecast: refers to the forecast for more than one year. Ultra-short-term load forecast and short-term load forecast are mainly used for air conditioning system control.

[0005] At present, the commonly used thermal prediction models are white box models represented by thermal balance equations, such as mechanism models. However, white box models often need to be accurately modeled according to the building's envelope structure, but since the model itself has a large number of parameters, such as the thermal capacitance and thermal conductivity of each sub-component, it is difficult to achieve accurate modeling.

[0006] There are two main types of data-driven black box models: neural network models and machine learning models. Neural network models include neural network models based on long short-term memory networks (LSTM), temporal convolutional networks (TCN), and multi-layer feedforward networks (BP). Machine learning models include machine learning models such as linear regression, autoregressive sliding average models, and support vector machines. Whether using neural network models or machine learning models for cold and hot load forecasting, basic steps such as data collection, data preprocessing, model construction, model training, and model prediction must be followed. Neural network prediction technology is widely used, but lacks reliable reasoning basis; machine learning models have stronger interpretability than neural network models. In general, the data collected by the data-driven black box model is the basis for subsequent reasoning, which determines the accuracy of reasoning, that is, the accuracy of prediction. However, due to the problem caused by the incompleteness of data collection, there is a lack of a solution that is suitable for intermittent heating / cooling scenarios with indoor temperature changes and can provide short-term and ultra-short-term load forecasting. Summary of the invention

[0007] The embodiment of the present invention provides an online prediction method for cooling and heating loads, which can be applied to intermittent heating / cooling scenarios with indoor temperature changes and can provide short-term and ultra-short-term load predictions. The method includes:

[0008] Obtaining a load forecast history record of a preset time length including the current moment, wherein the load forecast history record includes the energy supply load of the energy supply station, the outdoor environment data and the indoor temperature of the target building, and the energy supply load is a cooling load or a heating load;

[0009] Inputting the load forecast history records of a preset time length into a trained indoor temperature prediction model to predict the indoor temperature of the target building at the next moment, wherein the trained indoor temperature prediction model is obtained by constructing and training a neural network model with energy supply load, outdoor environmental parameters and historical indoor temperature as input variables and future indoor temperature as output variable;

[0010] Taking the predicted indoor temperature at the next moment as the expected value, reversely solve the energy load at the current moment;

[0011] According to the energy supply load at the current moment, the average values ​​of the inlet water temperature and the outlet water temperature of the energy supply system are determined.

[0012] The embodiment of the present invention further provides an online prediction device for cooling and heating loads, which can be applied to intermittent heating / cooling scenarios with indoor temperature changes and can provide short-term and ultra-short-term load predictions. The device includes:

[0013] A load forecast history acquisition module is used to obtain a load forecast history record of a preset time length including the current moment, wherein the load forecast history record includes the energy supply load of the energy supply station, the outdoor environment data and indoor temperature of the target building, and the historical indoor temperature data, wherein the energy supply load is a cooling load or a heating load;

[0014] A prediction module is used to input the load prediction history records of a preset time length into a trained indoor temperature prediction model to predict the indoor temperature of the target building at the next moment, wherein the trained indoor temperature prediction model is obtained by constructing and training a neural network model with energy supply load, outdoor environmental parameters and historical indoor temperature as input variables and future indoor temperature as output variable;

[0015] The energy load calculation module is used to reversely solve the energy load at the current moment by taking the predicted indoor temperature at the next moment as the expected value;

[0016] The water temperature determination module is used to determine the average value of the inlet water temperature and the outlet water temperature of the energy supply system according to the energy supply load at the current moment.

[0017] An embodiment of the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned online prediction method for cold and hot loads when executing the computer program.

[0018] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned online prediction method for cooling and heating loads is implemented.

[0019] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned online prediction method for cold and hot loads is implemented.

[0020] In an embodiment of the present invention, a load prediction history record of a preset time length including the current moment is obtained, the load prediction history record includes the energy load of the energy supply station, the outdoor environmental data and the indoor temperature of the target building, and the energy load is a cooling load or a heating load; the load prediction history record of the preset time length is input into a trained indoor temperature prediction model to predict the indoor temperature of the target building at the next moment, the trained indoor temperature prediction model is obtained by constructing a neural network model with the energy load, outdoor environmental parameters and historical indoor temperature as input variables and the future indoor temperature as output variable and training; the predicted indoor temperature at the next moment is taken as the expected value, and the energy load at the current moment is reversely solved; according to the energy load at the current moment, the average values ​​of the inlet and outlet water temperatures of the energy supply system are determined. In the above steps, an indoor temperature prediction model is constructed with energy load and outdoor environmental parameters as input variables and future indoor temperature as output variable, so that the constructed indoor temperature prediction model is more explanatory; in addition, the predicted indoor temperature at the next moment is taken as the expected value to reversely solve the energy load at the current moment. On the one hand, it considers the influence of the actual indoor temperature at the current moment on the cooling and heating load, and on the other hand, it considers the influence of the future indoor temperature (that is, taking the predicted indoor temperature at the next moment as the expected value) on the cooling and heating load. It can be applied to the short-term and ultra-short-term cooling and heating load prediction in intermittent heating / cooling scenarios with indoor temperature changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0022] Figure 1This is a flow chart of a method for online prediction of cold and hot loads in an embodiment of the present invention;

[0023] Figure 2 Detailed principle diagram of the online prediction method for cold and hot loads in an embodiment of the present invention;

[0024] Figure 3 This is a flow chart of energy supply load prediction at the current moment in an embodiment of the present invention;

[0025] Figure 4 This is a flow chart of energy supply load prediction at time t in an embodiment of the present invention;

[0026] Figure 5 Schematic diagram of an online prediction device for cold and hot loads in an embodiment of the present invention;

[0027] Figure 6 Schematic diagram of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0029] The inventors found that in the current load forecasting research, most load forecasts do not consider the impact of indoor temperature. Since the heat load data is a record of the current and past heat loads, if the indoor temperature is not considered, the recorded heat load value may be excessive (the indoor temperature is too high) or insufficient (the indoor temperature is too low) relative to the heat required by the building. In other words, the recorded heat load value is not the heat required by the building at a specified indoor temperature, and accurate heat load prediction cannot be achieved.

[0030] The present invention uses influencing factors such as outdoor temperature, atmospheric humidity, wind speed, solar irradiance, historical indoor temperature, etc. as independent variables, and uses the indoor temperature at the next moment as output. In fact, the purpose of the heating / cooling system is to make the indoor temperature reach the design standard value. The indoor temperature is the result of the accumulation of various factors such as outdoor temperature, atmospheric humidity, wind speed, solar irradiance, etc. over time. The cold / heat load prediction method with indoor temperature as the target is more explanatory.

[0031] Figure 1 This is a flow chart of a method for online prediction of cold and hot loads in an embodiment of the present invention, including:

[0032] Step 101, obtaining a load forecast history record of a preset time length including the current moment, wherein the load forecast history record includes the energy supply load of the energy supply station, the outdoor environment data and the indoor temperature of the target building, and the energy supply load is a cooling load or a heating load;

[0033] Step 102, inputting the load prediction history records of a preset time length into the trained indoor temperature prediction model to predict the indoor temperature of the target building at the next moment, wherein the trained indoor temperature prediction model is obtained by constructing and training a neural network model with energy supply load, outdoor environmental parameters and historical indoor temperature as input variables and future indoor temperature as output variable;

[0034] Step 103, using the predicted indoor temperature at the next moment as the expected value, reversely solve the energy supply load at the current moment;

[0035] Step 104, determining the average values ​​of the inlet water temperature and the outlet water temperature of the energy supply system according to the energy supply load at the current moment.

[0036] In an embodiment of the present invention, an indoor temperature prediction model is constructed with energy load, outdoor environmental parameters and historical indoor temperature as input variables and future indoor temperature as output variable, so that the constructed indoor temperature prediction model is more explanatory; in addition, the predicted indoor temperature at the next moment is taken as the expected value to reversely solve the energy load at the current moment, which considers the influence of the actual indoor temperature at the current moment on the cooling and heating load on the one hand, and the influence of the future indoor temperature (that is, taking the predicted indoor temperature at the next moment as the expected value) on the cooling and heating load on the other hand, and can be applicable to short-term and ultra-short-term cooling and heating load prediction in intermittent heating / cooling scenarios with indoor temperature changes.

[0037] Figure 2 This is a detailed schematic diagram of the online prediction method for cold and hot loads in an embodiment of the present invention, including four steps: indoor temperature prediction model construction, cold / hot load prediction, inlet and outlet water temperature mean prediction, and model self-learning. It should be noted that the models or model parameters applicable to the heating load and the cooling load are different, and they should be modeled and trained separately.

[0038] See also Figure 2 ,The steps for building the indoor temperature prediction model are given below.

[0039] In one embodiment, the method further comprises:

[0040] At first preset time intervals, the following steps are performed again to update the parameters of the indoor temperature prediction model:

[0041] Collect the indoor temperature and outdoor environment data of the target building in real time, and form a historical indoor temperature time series, a future indoor temperature time series and an outdoor environment data time series respectively, wherein the historical indoor temperature series is one time interval earlier than the future indoor temperature series;

[0042] Obtain the real-time energy supply load of the energy supply station and form the energy supply load time series;

[0043] Taking the energy load time series, outdoor environment data time series, and historical indoor temperature time series as input variables and the future indoor temperature time series as output variables, a neural network model is constructed as an indoor temperature prediction model.

[0044] The indoor temperature prediction model is trained using the energy load time series, outdoor environmental parameter time series, historical indoor temperature time series, and future indoor temperature series to determine the parameters of the indoor temperature prediction model and obtain a trained indoor temperature prediction model.

[0045] In one embodiment, the outdoor environment data includes at least one of dry-bulb temperature, wet-bulb temperature, solar irradiance, and wind speed.

[0046] In one embodiment, the method further comprises:

[0047] Real-time collection of water inlet temperature, water outlet temperature and flow rate of the main pipe on the user side of the energy supply station;

[0048] Calculate the real-time energy load based on the real-time inlet water temperature, outlet water temperature and flow rate.

[0049] The specific calculation method is:

[0050] Assume that the inlet water temperature at time t is The outlet water temperature is The real-time mass flow rate of the pipeline is m t , the specific heat capacity of water is c, the outlet water temperature is higher than the inlet water temperature during cooling, and the cooling load at time t is: When heating, the outlet water temperature is lower than the inlet water temperature, and the heat load at time t is:

[0051] In one embodiment, the indoor temperature prediction model adopts an LSTM machine learning model or a linear regression model.

[0052] In the above embodiment, when the indoor temperature prediction model is trained using the energy load time series, the outdoor environment parameter time series, the historical indoor temperature time series, and the future indoor temperature time series, all time series can be combined and organized into one record through data preprocessing and data normalization according to the requirements of the established indoor temperature prediction model. The data set is divided into a training set, a test set, and a validation set, and then the indoor temperature prediction model is trained and tested, and the parameters of the indoor temperature prediction model are finally confirmed through continuous iterations.

[0053] In the above embodiment, the process of updating the parameters of the indoor temperature prediction model every first preset time period is a self-learning process of the indoor temperature prediction model.

[0054] In step 102, the load prediction history records of a preset time length are input into the trained indoor temperature prediction model to predict the indoor temperature of the target building at the next moment.

[0055] Specifically, the heating / cooling load input at the current moment is used as a variable, combined with the outdoor dry-bulb temperature, wet-bulb temperature, solar irradiance, wind speed collected in real time at the current moment, and the current indoor temperature, as well as N-1 historical records of the heating / cooling load, outdoor dry-bulb temperature, wet-bulb temperature, solar irradiance, wind speed, historical indoor temperature, etc. before the current moment, which are combined into N data records (i.e., load prediction historical records of a preset time length) as input, and the indoor temperature at the next moment is predicted through the constructed indoor temperature prediction model.

[0056] In step 103, the predicted indoor temperature at the next moment is taken as the expected value, and the energy supply load at the current moment is reversely solved.

[0057] The solution method may adopt various solution methods such as traversal solution, gradient descent method, intelligent optimization algorithm, etc.

[0058] The embodiments of the present invention provide two situations of reversely solving the energy supply load.

[0059] The first one: energy load forecast at the current moment

[0060] In one embodiment, the predicted indoor temperature at the next moment is used as the expected value, and the energy supply load at the current moment is reversely solved, including:

[0061] The energy load at the current moment is calculated based on historical meteorological data, historical load data, meteorological forecast data, the expected value of the indoor temperature at the next moment and the real-time indoor temperature, wherein the real-time indoor temperature is measured after power control of the energy supply system is performed based on the energy load at the current moment.

[0062] Figure 3FIG. 1 is a flow chart of energy supply load prediction at the current moment in an embodiment of the present invention. Figure 3 As shown in the figure, the indoor temperature at the next moment can be measured because the indoor temperature can be measured by turning on the energy supply system (cooling / heating equipment) and injecting cold energy / heat energy into the room through the cooling / heating of the indoor heating / cooling equipment. At the same time, the heat generated inside the target building and the absorption or dissipation of heat outside jointly cause the change of indoor temperature. Therefore, if only one control cycle load forecast is performed, the following method can be used: Figure 3 The method shown.

[0063] The second type: energy supply load forecast for future control cycles

[0064] In one embodiment, the method further comprises:

[0065] Calculate the energy supply load at time t based on historical meteorological data, historical load data, meteorological forecast data, the expected value of indoor temperature at time t+1, and the expected value of indoor temperature at time t, where t is any time after the current time;

[0066] According to the energy supply load at time t, the average values ​​of the inlet water temperature and the outlet water temperature of the energy supply system are determined.

[0067] Figure 4 is a flow chart of energy supply load prediction at time t in an embodiment of the present invention, as shown in Figure 4 As shown, because the actual indoor temperature in the future control period (at time t) is unknown, the expected value of the indoor temperature at time t is used to replace the real-time detected indoor temperature at time t. Multiple iterations are performed to calculate the cooling and heating loads for a period of time in the future, that is, the energy supply load at time t.

[0068] In one embodiment, the method further comprises:

[0069] Every second preset time, the following steps are performed again to update the coefficients of the temperature fitting mathematical model:

[0070] According to different types of energy supply system terminals, a temperature fitting mathematical model between the energy supply load and the average value of the inlet water temperature and the outlet water temperature is constructed;

[0071] Fitting the coefficients of the temperature fitting mathematical model through the historical data of energy supply load and the historical data of the average values ​​of the inlet water temperature and the outlet water temperature;

[0072] According to the energy supply load at the current moment, determine the average values ​​of the inlet and outlet water temperatures of the energy supply system, including:

[0073] The type of the energy supply system terminal and the energy supply load at the current moment are input into the temperature fitting mathematical model to obtain the average values ​​of the inlet and outlet water temperatures of the energy supply system.

[0074] Types of energy supply systems include radiators and fan coils.

[0075] Radiator is the most commonly used equipment in heating system. There are many factors that affect the heat dissipation of radiator, such as the manufacturing condition of radiator and the use condition of radiator. In actual use, the heat dissipation of radiator is mainly affected by the inlet water temperature, outlet water temperature and indoor temperature of heat medium. The heat dissipation expression is:

[0076]

[0077] Where: Q s is the heat dissipation of the radiator, kW;

[0078] F is the heat dissipation area of ​​the radiator, m2;

[0079] t1, t2 are the water inlet and outlet temperatures of the radiator, °C;

[0080] t n is the indoor air temperature, °C;

[0081] a and b are coefficients determined by experiments.

[0082] For fan coil units, the heat dissipation is mainly affected by the inlet water temperature, outlet water temperature, water flow, wind speed and indoor temperature. The heat dissipation expression is:

[0083]

[0084] Where: Q p is the heat dissipation of the fan coil, kW;

[0085] a, b, c, d are the heat dissipation characteristic parameters of the fan coil unit;

[0086] V y is the heat exchange wind speed of the fan coil, m / s;

[0087] V w is the water velocity in the coil, m / s;

[0088] F is the coil area, m2;

[0089] t1, t2 are the inlet and outlet water temperatures of the fan coil, °C;

[0090] t n is the indoor air temperature, ℃.

[0091] In the above embodiment, the above steps are executed again every second preset time period to update the coefficients of the temperature fitting mathematical model as a self-learning process.

[0092] In summary, the method proposed in the embodiment of the present invention constructs a cooling and heating load prediction model with indoor temperature as output. On the one hand, it considers the influence of the actual indoor temperature at the current moment on the cooling and heating load, and on the other hand, it considers the influence of the future indoor design temperature on the cooling and heating load. It can be applied to short-term and ultra-short-term cooling and heating load prediction in intermittent heating / cooling scenarios with indoor temperature changes, and at the same time has the self-learning and self-correction functions of the cooling and heating load prediction model.

[0093] The embodiment of the present invention further provides an online prediction device for cold and hot loads, the principle of which is similar to the online prediction method for cold and hot loads, and will not be described in detail here.

[0094] Figure 5 This is a schematic diagram of an online prediction device for cold and hot loads in an embodiment of the present invention, comprising:

[0095] A load forecast history acquisition module 501 is used to obtain a load forecast history record of a preset time length including the current moment, wherein the load forecast history record includes the energy supply load of the energy supply station, the outdoor environment data and the indoor temperature of the target building, and the energy supply load is a cooling load or a heating load;

[0096] Prediction module 502, used to input the load prediction history records of a preset time length into the trained indoor temperature prediction model, and predict the indoor temperature of the target building at the next moment, wherein the trained indoor temperature prediction model is obtained by constructing and training a neural network model with energy supply load, outdoor environmental parameters and historical indoor temperature as input variables and future indoor temperature as output variable;

[0097] The energy supply load calculation module 503 is used to reversely solve the energy supply load at the current moment by taking the predicted indoor temperature at the next moment as the expected value;

[0098] The water temperature determination module 504 is used to determine the average value of the inlet water temperature and the outlet water temperature of the energy supply system according to the energy supply load at the current moment.

[0099] In one embodiment, the outdoor environment data includes at least one of dry-bulb temperature, wet-bulb temperature, solar irradiance, and wind speed.

[0100] In one embodiment, the device further includes an indoor temperature prediction model building module 505, which is used to:

[0101] At first preset time intervals, the following steps are performed again to update the parameters of the indoor temperature prediction model:

[0102] Collect the indoor temperature and outdoor environment data of the target building in real time, and form a historical indoor temperature time series, a future indoor temperature time series and an outdoor environment data time series respectively, wherein the historical indoor temperature series is one time interval earlier than the future indoor temperature series;

[0103] Obtain the real-time energy supply load of the energy supply station and form the energy supply load time series;

[0104] Taking the energy load time series, outdoor environment data time series, and historical indoor temperature series as input variables and the future indoor temperature time series as output variables, a neural network model is constructed as an indoor temperature prediction model.

[0105] The indoor temperature prediction model is trained using the energy load time series, outdoor environmental parameter time series, historical indoor temperature time series, and future indoor temperature time series to determine the parameters of the indoor temperature prediction model and obtain a trained indoor temperature prediction model.

[0106] In one embodiment, the indoor temperature prediction model adopts an LSTM machine learning model or a linear regression model.

[0107] In one embodiment, the indoor temperature prediction model building module is further used to:

[0108] Real-time collection of water inlet temperature, water outlet temperature and flow rate of the main pipe on the user side of the energy supply station;

[0109] Calculate the real-time energy load based on the real-time inlet water temperature, outlet water temperature and flow rate.

[0110] In one embodiment, the energy supply load calculation module 503 is used to:

[0111] The energy load at the current moment is calculated based on historical meteorological data, historical load data, meteorological forecast data, the expected value of the indoor temperature at the next moment and the real-time indoor temperature, wherein the real-time indoor temperature is measured after power control of the energy supply system is performed based on the energy load at the current moment.

[0112] In one embodiment, the energy supply load calculation module 503 is further used to:

[0113] Calculate the energy supply load at time t based on historical meteorological data, historical load data, meteorological forecast data, the expected value of indoor temperature at time t+1, and the expected value of indoor temperature at time t, where t is any time after the current time;

[0114] According to the energy supply load at time t, the average values ​​of the inlet water temperature and the outlet water temperature of the energy supply system are determined.

[0115] In one embodiment, the device further includes a temperature fitting mathematical model building module 506, which is used to:

[0116] Every second preset time, the following steps are performed again to update the coefficients of the temperature fitting mathematical model:

[0117] According to different types of energy supply system terminals, a temperature fitting mathematical model between the energy supply load and the average value of the inlet water temperature and the outlet water temperature is constructed;

[0118] Fitting the coefficients of the temperature fitting mathematical model through the historical data of energy supply load and the historical data of the average values ​​of the inlet water temperature and the outlet water temperature;

[0119] According to the current energy supply load, determine the average values ​​of the inlet and outlet water temperatures of the energy supply system, including:

[0120] The type of the energy supply system terminal and the energy supply load at the current moment are input into the temperature fitting mathematical model to obtain the average values ​​of the inlet and outlet water temperatures of the energy supply system.

[0121] To sum up, in the device proposed in the embodiment of the present invention, a heating and cooling load prediction model is constructed with indoor temperature as output. On the one hand, the influence of the actual indoor temperature at the current moment on the heating and cooling load is considered, and on the other hand, the influence of the future indoor design temperature on the heating and cooling load is considered. It can be suitable for short-term and ultra-short-term heating and cooling load prediction in intermittent heating / cooling scenarios with indoor temperature changes, and at the same time has the self-learning and self-correction functions of the heating and cooling load prediction model.

[0122] An embodiment of the present invention further provides a computer device, Figure 6 Schematic diagram of a computer device in an embodiment of the present invention, the computer device 600 includes a memory 610, a processor 620, and a computer program 630 stored in the memory 610 and executable on the processor 620, and the processor 620 implements the above-mentioned online prediction method for cold and hot loads when executing the computer program 630.

[0123] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned online prediction method for cooling and heating loads is implemented.

[0124] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned online prediction method for cold and hot loads is implemented.

[0125] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0126] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0127] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0128] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0129] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for online prediction of cold and hot loads, characterized in that: include: Obtaining a load forecast history record of a preset time length including the current moment, wherein the load forecast history record includes the energy supply load of the energy supply station, the outdoor environment data and the indoor temperature of the target building, and the energy supply load is a cooling load or a heating load; Inputting the load forecast history records of a preset time length into a trained indoor temperature prediction model to predict the indoor temperature of the target building at the next moment, wherein the trained indoor temperature prediction model is obtained by constructing and training a neural network model with energy supply load, outdoor environmental parameters and historical indoor temperature as input variables and future indoor temperature as output variable; Taking the predicted indoor temperature at the next moment as the expected value, reversely solve the energy load at the current moment; According to the energy supply load at the current moment, the average values ​​of the inlet water temperature and the outlet water temperature of the energy supply system are determined.

2. The method according to claim 1, characterized in that The outdoor environment data includes at least one of dry-bulb temperature, wet-bulb temperature, solar irradiance, and wind speed.

3. The method according to claim 1, characterized in that Also includes: At first preset time intervals, the following steps are performed again to update the parameters of the indoor temperature prediction model: Collect the indoor temperature and outdoor environment data of the target building in real time, and form a historical indoor temperature time series, a future indoor temperature time series and an outdoor environment data time series respectively, wherein the historical indoor temperature series is one time interval earlier than the future indoor temperature series; Obtain the real-time energy supply load of the energy supply station and form the energy supply load time series; Taking the energy load time series, outdoor environment data time series, and historical indoor temperature series as input variables and the future indoor temperature time series as output variables, a neural network model is constructed as an indoor temperature prediction model. The indoor temperature prediction model is trained using the energy load time series, outdoor environmental parameter time series, historical indoor temperature time series, and future indoor temperature time series to determine the parameters of the indoor temperature prediction model and obtain a trained indoor temperature prediction model.

4. The method according to claim 3, characterized in that Also includes: Real-time collection of water inlet temperature, water outlet temperature and flow rate of the main pipe on the user side of the energy supply station; Calculate the real-time energy load based on the real-time inlet water temperature, outlet water temperature and flow rate.

5. The method according to claim 1, characterized in that The indoor temperature prediction model uses an LSTM machine learning model or a linear regression model.

6. The method according to claim 1, characterized in that Taking the predicted indoor temperature at the next moment as the expected value, reversely solve the energy load at the current moment, including: The energy load at the current moment is calculated based on historical meteorological data, historical load data, meteorological forecast data, the expected value of the indoor temperature at the next moment and the real-time indoor temperature, wherein the real-time indoor temperature is measured after power control of the energy supply system is performed based on the energy load at the current moment.

7. The method according to claim 6, characterized in that Also includes: Calculate the energy supply load at time t based on historical meteorological data, historical load data, meteorological forecast data, the expected value of indoor temperature at time t+1, and the expected value of indoor temperature at time t, where t is any time after the current time; According to the energy supply load at time t, the average values ​​of the inlet water temperature and the outlet water temperature of the energy supply system are determined.

8. The method according to claim 1, characterized in that Also includes: Every second preset time, the following steps are performed again to update the coefficients of the temperature fitting mathematical model: According to different types of energy supply system terminals, a temperature fitting mathematical model between the energy supply load and the average value of the inlet water temperature and the outlet water temperature is constructed; Fitting the coefficients of the temperature fitting mathematical model through the historical data of energy supply load and the historical data of the average values ​​of the inlet water temperature and the outlet water temperature; According to the current energy supply load, determine the average values ​​of the inlet and outlet water temperatures of the energy supply system, including: The type of the energy supply system terminal and the energy supply load at the current moment are input into the temperature fitting mathematical model to obtain the average values ​​of the inlet and outlet water temperatures of the energy supply system.

9. An online prediction device for cold and hot loads, characterized in that: include: A load forecast history acquisition module, used to obtain a load forecast history record of a preset time length including the current moment, wherein the load forecast history record includes the energy supply load of the energy supply station, the outdoor environment data and the indoor temperature of the target building, and the energy supply load is a cooling load or a heating load; A prediction module is used to input the load prediction history records of a preset time length into a trained indoor temperature prediction model to predict the indoor temperature of the target building at the next moment, wherein the trained indoor temperature prediction model is obtained by constructing and training a neural network model with energy supply load, outdoor environmental parameters and historical indoor temperature as input variables and future indoor temperature as output variable; The energy load calculation module is used to reversely solve the energy load at the current moment by taking the predicted indoor temperature at the next moment as the expected value; The water temperature determination module is used to determine the average value of the inlet water temperature and the outlet water temperature of the energy supply system according to the energy supply load at the current moment.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

12. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Citation Information

Cited By

  • Building cooling and heating load prediction method based on generative adversarial network

    CN119398281A

  • Indoor temperature determination method fusing physical constraints

    CN121859758A