Method and System for Predicting and Adaptively Correcting End Load of Ground Source Heat Pump System

Abstract

The present invention relates to the field of energy management and intelligent control technologies, and provides a method and system for predicting and adaptively correcting the end load of a ground source heat pump system. The method includes: constructing a day-ahead load prediction model to calculate the day-ahead load prediction value; constructing a day-ahead load prediction deviation correction model to obtain the day-ahead load prediction deviation correction value; obtaining a day-ahead load prediction value vector based on the day-ahead load prediction deviation correction value; constructing an intraday load prediction model to calculate the intraday load prediction value; respectively calculating the credibility of the predicted values at the corresponding moments in the intraday load prediction value and the day-ahead load prediction value vector and the historical measured load data, and calculating the predicted load issued value based on the predicted values at the corresponding moments in the intraday load prediction value and the day-ahead load prediction value vector and the credibility. By constructing a double-layer load prediction model combined with a meteorological parameter correction model, the present invention improves the adaptability of the system to dynamic environmental changes and the accuracy of load prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of energy management and intelligent control, and particularly relates to a method and system for predicting and adaptively correcting the end - load of a ground - source heat pump system. Background Art

[0002] With the intensification of the global energy crisis and environmental protection awareness, improving energy utilization efficiency has become an important goal of social development. As a key energy facility in heating, air - conditioning, and industrial processes, the operating state of the cold - heat source system directly affects the energy efficiency and operating cost of the entire energy management system. However, due to the dynamic changes of external factors such as environmental temperature and humidity, as well as the random fluctuations of load demands, traditional fixed scheduling strategies are often difficult to adapt to the actual operating environment, resulting in energy waste and unstable equipment operation.

[0003] Load prediction is the core link of the optimal scheduling of the cold - heat source system. Its goal is to predict the load demand of the system in advance based on historical operation data and future meteorological parameters. The mainstream methods include those based on statistical models (such as time - series analysis and regression modeling) and intelligent algorithms (such as neural networks and support vector machines). However, single models often show limitations when dealing with complex non - linear characteristics or scarce data. Summary of the Invention

[0004] The purpose of the present invention is to solve at least one technical problem in the background art, and to provide a method and system for predicting and adaptively correcting the end - load of a ground - source heat pump system.

[0005] To achieve the above - mentioned purpose, the present invention provides a method for predicting and adaptively correcting the end - load of a ground - source heat pump system, including:

[0006] Obtaining historical operation load data, historical predicted load data, historical measured load data, and historical outdoor predicted meteorological data;

[0007] Constructing a day - ahead load prediction model based on the historical operation load data and the historical measured load data, and calculating a day - ahead load prediction value based on the day - ahead load prediction model;

[0008] Constructing a day - ahead load prediction deviation correction model based on the historical predicted load data, the historical measured load data, and the historical outdoor predicted meteorological data, and calculating a day - ahead load prediction deviation correction value based on the day - ahead load prediction deviation correction model;

[0009] Obtaining a day - ahead load prediction value vector based on the day - ahead load prediction deviation correction value;

[0010] Constructing an intraday load prediction model based on the historical operation load data, the historical predicted load data, and the historical outdoor predicted meteorological data, and calculating an intraday load prediction value based on the intraday load prediction model;

[0011] Calculate the credibility of the predicted values at the corresponding moments in the intraday load prediction value and the day-ahead load prediction value vector and the historical measured load data respectively, and calculate the issued value of the predicted load based on the predicted values and credibility at the corresponding moments in the intraday load prediction value and the day-ahead load prediction value vector.

[0012] According to one aspect of the present invention, constructing a day-ahead load prediction model based on historical operating load data and historical measured load data, and calculating a day-ahead load prediction value based on the day-ahead load prediction model includes:

[0013] Use historical operating load data of at least the past two weeks to construct a Holt-Winters additive model as the day-ahead load prediction model for basic prediction of day-ahead load prediction:

[0014] ;

[0015] In the formula: is the basic value of the day-ahead load prediction at the moment; is the load value at the moment; is the smoothed value at the moment; is the trend value at the moment; is the periodic value at the moment; is the cycle length, which is taken as 24 in the day-ahead load prediction model; is the smoothing coefficient, ; is the trend coefficient, ; is the seasonal coefficient, ;

[0016] Predict the day-ahead load prediction value for the future moment: .

[0017] According to one aspect of the present invention, constructing a day-ahead load prediction deviation correction model based on historical predicted load data, historical measured load data, and historical outdoor predicted meteorological data, and calculating a day-ahead load prediction deviation correction value based on the day-ahead load prediction deviation correction model includes:

[0018] Calculate the residual using historical measured load data and historical predicted load data:

[0019] ;

[0020] In the formula:

[0021] is the load forecasting residual at the previous moment ; is the load value at the th moment; is the previous-day load forecast value at the th moment;

[0022] Adopt the method of multiple linear regression, and use the predicted outdoor temperature and the predicted outdoor relative humidity in the historical outdoor forecast meteorological data of at least the past two weeks, as well as the load forecasting residual to construct a previous-day load forecasting deviation correction model:

[0023] ;

[0024] After training, obtain the parameters of the previous-day load forecasting deviation correction model;

[0025] For the predicted outdoor temperature and the predicted outdoor relative humidity at the future th moment, substitute them into the trained previous-day load forecasting deviation correction model:

[0026] ;

[0027] The previous-day load forecasting deviation correction value is:

[0028] .

[0029] According to one aspect of the present invention, the previous-day load forecast value vector is obtained based on the previous-day load forecasting deviation correction value as:

[0030] Based on the previous-day load forecasting deviation correction value , where, taking as 0 - 23, the previous-day load forecast value vector for the next 24 hours can be obtained:

[0031] .

[0032] According to one aspect of the present invention, constructing an intraday load forecasting model based on the historical operating load data, historical forecast load data, and historical outdoor forecast meteorological data, and calculating the intraday load forecast value based on the intraday load forecasting model includes:

[0033] Adopt the least squares method, and use the historical measured load data, historical forecast load data, and historical outdoor forecast meteorological data of at least the past two weeks to construct an intraday load forecasting model:

[0034] ;

[0035] Wherein: is the predicted value of the in - day load at the th moment; is the number of hours considered for historical load; is the historical load coefficient, ; is the load value at the th past moment; is the number of hours considered for historical residual; is the historical residual coefficient, ; is the residual at the th past moment; is the vector of meteorological parameter coefficients; C is the constant term coefficient.

[0036] According to one aspect of the present invention, the credibility of the predicted values corresponding to the in - day load prediction value and the day - ahead load prediction value vector at the corresponding moment is calculated respectively based on the in - day load prediction value, the predicted value at the corresponding moment in the day - ahead load prediction value vector and the historical measured load data, and the predicted load issued value is calculated based on the in - day load prediction value, the predicted value at the corresponding moment in the day - ahead load prediction value vector and the credibility, including:

[0037] Extract the in - day load prediction value and the predicted value at the corresponding moment of the day - ahead load prediction value vector , and calculate the credibility of the in - day load prediction value , the predicted value at the corresponding moment of the day - ahead load prediction value vector and the historical measured load value , :

[0038] ;

[0039] ;

[0040] At the next moment, the formula for calculating the predicted load issued value issued is as follows:

[0041] .

[0042] Furthermore, to achieve the above object, the present invention also provides a ground - source heat pump system end - load prediction and adaptive correction system, including:

[0043] A data acquisition module, which acquires historical operating load data, historical predicted load data, historical measured load data, and historical outdoor predicted meteorological data;

[0044] The day-ahead load forecasting model construction module constructs a day-ahead load forecasting model based on historical operating load data and historical measured load data, and calculates a day-ahead load forecasting value based on the day-ahead load forecasting model;

[0045] The day-ahead load forecasting deviation correction model construction module constructs a day-ahead load forecasting deviation correction model based on historical forecast load data, historical measured load data, and historical outdoor forecast meteorological data, and calculates a day-ahead load forecasting deviation correction value based on the day-ahead load forecasting deviation correction model;

[0046] The day-ahead load forecasting value vector acquisition module obtains a day-ahead load forecasting value vector based on the day-ahead load forecasting deviation correction value;

[0047] The intra-day load forecasting model construction module constructs an intra-day load forecasting model based on historical operating load data, historical forecast load data, and historical outdoor forecast meteorological data, and calculates an intra-day load forecasting value based on the intra-day load forecasting model;

[0048] The predicted load transmission value acquisition module calculates the credibility of the predicted values at the corresponding times in the intra-day load forecasting value and the day-ahead load forecasting value vector with respect to the historical measured load data, and calculates a predicted load transmission value based on the intra-day load forecasting value, the predicted values at the corresponding times in the day-ahead load forecasting value vector, and the credibility.

[0049] Furthermore, to achieve the above object, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, the above-mentioned ground source heat pump system end load forecasting and adaptive correction method is implemented.

[0050] Furthermore, to achieve the above object, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned ground source heat pump system end load forecasting and adaptive correction method is implemented.

[0051] According to the solution of the present invention, by introducing meteorological variables such as environmental temperature and humidity, the present invention constructs a regression model to dynamically correct the forecasting results, which helps to improve the forecasting accuracy. At the same time, the real-time update and rolling training strategy of the model parameters can ensure that the model adapts to the dynamic changes of the operating environment and enhance the system robustness.

[0052] To achieve closed-loop operation, it is necessary to combine the forecasting credibility evaluation and feedback control mechanism to evaluate the model performance in real time and adjust the forecasting strategy. This solution can not only improve the load forecasting accuracy, reduce energy waste, but also significantly reduce the operating cost and equipment loss, and contribute to the development of green energy.

[0053] According to the solution of the present invention, by constructing a two - layer load forecasting model, the present invention effectively improves the load forecasting accuracy and scheduling decision - making level of the cold and heat source system. Its day - ahead forecasting model captures long - term trends and seasonal variations, and the intra - day forecasting model deals with short - term fluctuations. Combined with the meteorological parameter correction model, it enhances the adaptability of the system to dynamic environmental changes. Through the rolling update and feedback control mechanism, the system can automatically adjust the model parameters to ensure the stability and continuous optimization of the forecasting results.

[0054] In actual operation, the present invention can significantly reduce energy waste and equipment wear, lower the operating cost, and improve the reliability and economic benefits of the system operation. In addition, through the dynamic optimization of energy distribution, the system effectively reduces carbon emissions and environmental pollution, and has important social and environmental values. At the same time, the invention has good versatility for various energy systems and can be widely applied to scenarios such as regional energy management, intelligent building control, and industrial process optimization, providing technical support for the comprehensive upgrade of intelligent energy management systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 Schematically showing the flowchart of the end - load forecasting and adaptive correction method for a ground - source heat pump system according to an embodiment of the present invention;

[0056] Figure 2 Schematically showing the operating principle diagram of the end - load forecasting and adaptive correction method for a ground - source heat pump system according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] Now, the content of the present invention will be described with reference to exemplary embodiments. It should be understood that the described embodiments are only for enabling those of ordinary skill in the art to better understand and thus implement the content of the present invention, rather than implying any limitation to the scope of the present invention.

[0058] As used herein, the term "comprising" and its variants are to be construed as open - ended terms meaning "including but not limited to". The term "based on" is to be construed as "at least partially based on". The terms "an embodiment" and "one embodiment" are to be construed as "at least one embodiment".

[0059] Figure 1 Schematically showing the flowchart of the end - load forecasting and adaptive correction method for a ground - source heat pump system according to an embodiment of the present invention. As Figure 1 shown, in this embodiment, the end - load forecasting and adaptive correction method for a ground - source heat pump system includes:

[0060] Obtain historical operating load data, historical forecast load data, historical measured load data, and historical outdoor forecast meteorological data (outdoor temperature, outdoor relative humidity);

[0061] Construct a day-ahead load forecasting model based on historical operating load data and historical measured load data, and calculate the day-ahead load forecast value based on the day-ahead load forecasting model;

[0062] Construct a day-ahead load forecasting deviation correction model based on historical forecast load data, historical measured load data, and historical outdoor forecast meteorological data, and calculate the day-ahead load forecasting deviation correction value based on the day-ahead load forecasting deviation correction model;

[0063] Obtain the day-ahead load forecast value vector based on the day-ahead load forecasting deviation correction value;

[0064] Construct an intraday load forecasting model based on historical operating load data, historical forecast load data, and historical outdoor forecast meteorological data, and calculate the intraday load forecast value based on the intraday load forecasting model;

[0065] Calculate the credibility of the forecast values at the corresponding times in the intraday load forecast value and the day-ahead load forecast value vector with respect to the historical measured load data, and calculate the forecast load issued value based on the intraday load forecast value, the forecast values at the corresponding times in the day-ahead load forecast value vector, and the credibility.

[0066] In this embodiment, the above-obtained data will be updated as the system and the forecasting model run.

[0067] Further, according to an embodiment of the present invention, constructing a day-ahead load forecasting model based on historical operating load data and historical measured load data, and calculating the day-ahead load forecast value based on the day-ahead load forecasting model includes:

[0068] Use the historical operating load data of at least the past two weeks to construct a Holt-Winters additive model as the day-ahead load forecasting model for basic forecasting of day-ahead load forecasting:

[0069] ;

[0070] In the formula: is the basic value of the day-ahead load forecast at the th moment; is the load value at the th moment; is the smoothing value at the th moment; is the trend value at the th moment; is the periodic value at the th moment; is the cycle length, which is taken as 24 in the day-ahead load forecasting model; is the smoothing coefficient, ; is the trend coefficient, ; is the seasonal coefficient, ;

[0071] For the future time, the day-ahead load forecast value is obtained by prediction: .

[0072] Furthermore, according to an embodiment of the present invention, a day-ahead load forecast deviation correction model is constructed based on historical forecast load data, historical measured load data, and historical outdoor forecast meteorological data, and the day-ahead load forecast deviation correction value is calculated based on the day-ahead load forecast deviation correction model, including:

[0073] Calculate the residual using historical measured load data and historical forecast load data:

[0074] ;

[0075] In the formula:

[0076] is the load forecast residual at the th hour of the day-ahead; is the load value at the th hour; is the day-ahead load forecast value at the th hour;

[0077] Adopt the method of multiple linear regression, and use the predicted outdoor temperature and predicted outdoor relative humidity in the historical outdoor forecast meteorological data of at least the past two weeks, and the load forecast residual to construct a day-ahead load forecast deviation correction model:

[0078] ;

[0079] After training, obtain the day-ahead load forecast deviation correction model parameters ;

[0080] For the predicted outdoor temperature and predicted outdoor relative humidity at the future time, substitute them into the trained day-ahead load forecast deviation correction model:

[0081] ;

[0082] The day-ahead load forecast deviation correction value is:

[0083] .

[0084] Further, according to an embodiment of the present invention, the day-ahead load forecast value vector is obtained based on the day-ahead load forecast deviation correction value as follows:

[0085] Based on the day-ahead load forecast deviation correction value , where, taking as 0 - 23, the day-ahead load forecast value vector for the next 24 hours can be obtained :

[0086] .

[0087] Further, according to an embodiment of the present invention, an intraday load forecast model is constructed based on historical operating load data, historical forecast load data, and historical outdoor forecast meteorological data, and the intraday load forecast value is calculated based on the intraday load forecast model, including:

[0088] Using the least squares method, an intraday load forecast model is constructed using at least historical measured load data, historical forecast load data, and historical outdoor forecast meteorological data from the past two weeks:

[0089] ;

[0090] In the formula: is the intraday load forecast value at the moment; is the number of hours considered for historical load; is the historical load coefficient, ; is the load value at the th moment in the past; is the number of hours considered for historical residual; is the historical residual coefficient, ; is the residual at the th moment in the past; is the vector of meteorological parameter coefficients; C is the constant term coefficient.

[0091] Further, according to an embodiment of the present invention, the credibility of the predicted values at the corresponding moments in the intraday load forecast value and the day-ahead load forecast value vector is calculated respectively, and the predicted load distribution value is calculated based on the intraday load forecast value, the predicted values at the corresponding moments in the day-ahead load forecast value vector, and the credibility, including:

[0092] Extract the intraday load forecast value and the predicted value at the corresponding moment of the day-ahead load forecast value vector , and calculate the intraday load forecast value , the predicted value at the corresponding moment of the day-ahead load forecast value vector and the historical measured load value credibility 、 :

[0093] ;

[0094] ;

[0095] At the next moment, the predicted load value to be issued The calculation formula is as follows:

[0096] 。

[0097] In this embodiment, the day-ahead load prediction value vector obtains the vector of the load prediction values within the next day, corresponding to the predicted loads at 24 moments respectively. For example, when predicting the load at 10 o'clock, it is necessary to extract the load prediction value corresponding to 10 o'clock in the day-ahead load prediction vector of the current day and the in-day load prediction value to perform the credibility calculation together.

[0098] According to the above solution of the present invention, based on the day-ahead load prediction model and the in-day load prediction model, combined with the measured load data, the end-load prediction and adaptive correction method of the ground source heat pump system of the present invention is constructed. As Figure 2 shown, the specific operation steps of this method are as follows:

[0099] Step 1. Update the model coefficients of the day-ahead load prediction model and the day-ahead load prediction deviation correction model, and calculate the day-ahead load prediction value vector within one day 。

[0100] Step 2. Update the model coefficients of the in-day load prediction model, and calculate the in-day load prediction value at the prediction moment 。

[0101] Step 3. Calculate the credibility of the prediction value at a historical moment 、 , and calculate the predicted load value to be issued 。

[0102] Step 4. Issue the predicted load value to be issued , and after guiding the actual operation of the system, obtain the actual operation load , and update the historical measured load data set; update the data set of the historical predicted load data according to the day-ahead and in-day load prediction values; update the data set of the historical predicted meteorological data from the historical outdoor predicted meteorological data.

[0103] Step 5. If the current day's operation (prediction) has not ended, return to Step 2 for intraday load prediction; if the operation has ended, determine whether to continue the operation for the next day. If so, obtain the predicted meteorological data for the next day, and then return to Step 1 to loop; otherwise, end the operation.

[0104] According to the above solution of the present invention, by introducing meteorological variables such as ambient temperature and humidity, the present invention constructs a regression model to dynamically correct the prediction results, which helps to improve the prediction accuracy. At the same time, the real-time update of model parameters and the rolling training strategy can ensure that the model adapts to the dynamic changes of the operating environment and enhance the system robustness.

[0105] To achieve closed-loop operation, it is necessary to combine the prediction credibility evaluation and feedback control mechanism to evaluate the model performance in real time and adjust the prediction strategy. This solution can not only improve the load prediction accuracy, reduce energy waste, but also significantly reduce the operating cost and equipment loss, and contribute to the development of green energy.

[0106] According to the above solution of the present invention, by constructing a two-layer load prediction model, the present invention effectively improves the load prediction accuracy and dispatching decision-making level of the cold and heat source system. Its day-ahead prediction model captures the long-term trend and seasonal changes, and the intraday prediction model responds to short-term fluctuations. Combining with the meteorological parameter correction model, it enhances the system's adaptability to dynamic environmental changes. Through the rolling update and feedback control mechanism, the system can automatically adjust the model parameters to ensure the stability and continuous optimization of the prediction results.

[0107] In actual operation, the present invention can significantly reduce energy waste and equipment loss, reduce the operating cost, and improve the reliability and economic benefits of system operation. In addition, through the dynamic optimization of energy distribution, the system effectively reduces carbon emissions and environmental pollution, and has important social and environmental values. At the same time, the invention has good versatility for various energy systems and can be widely applied to scenarios such as regional energy management, intelligent building control, and industrial process optimization, providing technical support for the comprehensive upgrade of intelligent energy management systems.

[0108] Furthermore, to achieve the above object, the present invention also provides a ground source heat pump system terminal load prediction and adaptive correction system, including:

[0109] A data acquisition module, which acquires historical operation load data, historical predicted load data, historical measured load data, and historical outdoor predicted meteorological data;

[0110] A day-ahead load prediction model construction module, which constructs a day-ahead load prediction model based on historical operation load data and historical measured load data, and calculates the day-ahead load prediction value based on the day-ahead load prediction model;

[0111] The day-ahead load prediction deviation correction model construction module constructs a day-ahead load prediction deviation correction model based on historical predicted load data, historical measured load data, and historical outdoor predicted meteorological data, and calculates a day-ahead load prediction deviation correction value based on the day-ahead load prediction deviation correction model;

[0112] The day-ahead load prediction value vector acquisition module obtains a day-ahead load prediction value vector based on the day-ahead load prediction deviation correction value;

[0113] The intra-day load prediction model construction module constructs an intra-day load prediction model based on historical operating load data, historical predicted load data, and historical outdoor predicted meteorological data, and calculates an intra-day load prediction value based on the intra-day load prediction model;

[0114] The predicted load dispatch value acquisition module calculates the credibility of the predicted values at corresponding times in the intra-day load prediction value and the day-ahead load prediction value vector with respect to the historical measured load data, and calculates a predicted load dispatch value based on the intra-day load prediction value, the predicted values at corresponding times in the day-ahead load prediction value vector, and the credibility.

[0115] In this embodiment, the above-obtained various data will be updated as the system and the prediction model run.

[0116] Further, according to an embodiment of the present invention, constructing a day-ahead load prediction model based on historical operating load data and historical measured load data, and calculating a day-ahead load prediction value based on the day-ahead load prediction model, includes:

[0117] Using historical operating load data for at least the past two weeks, constructing a Holt-Winters additive model as the day-ahead load prediction model for basic prediction of day-ahead load prediction:

[0118] ;

[0119] Where: is the basic value of the day-ahead load prediction at the th moment; is the load value at the th moment; is the smoothing value at the th moment; is the trend value at the th moment; is the periodic value at the th moment; is the cycle length, which is taken as 24 in the day-ahead load prediction model; is the smoothing coefficient, ; is the trend coefficient, ; is the seasonal coefficient, ;

[0120] For the future time, the day-ahead load forecast value is obtained by prediction: .

[0121] Furthermore, according to an embodiment of the present invention, a day-ahead load forecast deviation correction model is constructed based on historical forecast load data, historical measured load data, and historical outdoor forecast meteorological data, and the day-ahead load forecast deviation correction value is calculated based on the day-ahead load forecast deviation correction model, including:

[0122] The residual is calculated using historical measured load data and historical forecast load data:

[0123] ;

[0124] In the formula:

[0125] is the load forecast residual at the th time of the day-ahead; is the load value at the th time; is the day-ahead load forecast value at the th time;

[0126] Using the method of multiple linear regression, the predicted outdoor temperature and the predicted outdoor relative humidity in the historical outdoor forecast meteorological data of at least the past two weeks, and the load forecast residual are used to construct a day-ahead load forecast deviation correction model:

[0127] ;

[0128] After training, the parameters of the day-ahead load forecast deviation correction model are obtained;

[0129] For the predicted outdoor temperature and the predicted outdoor relative humidity at the future time, substitute them into the trained day-ahead load forecast deviation correction model:

[0130] ;

[0131] The day-ahead load forecast deviation correction value is:

[0132] .

[0133] Furthermore, according to an embodiment of the present invention, the day-ahead load forecast value vector is obtained based on the day-ahead load forecast deviation correction value as:

[0134] Based on the correction value of the day-ahead load prediction deviation , where, taking as 0 - 23, the day-ahead load prediction value vector for the next 24 hours can be obtained :

[0135] .

[0136] Furthermore, according to an embodiment of the present invention, an intraday load prediction model is constructed based on historical operating load data, historical predicted load data, and historical outdoor predicted meteorological data, and the intraday load prediction value is calculated based on the intraday load prediction model, including:

[0137] Using the least squares method, an intraday load prediction model is constructed using at least the historical measured load data, historical predicted load data, and historical outdoor predicted meteorological data of the past two weeks:

[0138] ;

[0139] In the formula: is the intraday load prediction value at the th moment; is the number of hours considered for historical load; is the historical load coefficient, ; is the load value at the th moment in the past; is the number of hours considered for historical residual; is the historical residual coefficient, ; is the residual at the th moment in the past; is the vector of meteorological parameter coefficients; C is the constant term coefficient.

[0140] Furthermore, according to an embodiment of the present invention, the credibility of the predicted values at the corresponding moments in the intraday load prediction value and the day-ahead load prediction value vector with respect to the historical measured load data is calculated respectively, and the predicted load issuance value is calculated based on the intraday load prediction value, the predicted values at the corresponding moments in the day-ahead load prediction value vector, and the credibility, including:

[0141] Extract the intraday load prediction value and the predicted value at the corresponding moment of the day-ahead load prediction value vector , and calculate the credibility of the intraday load prediction value , the predicted value at the corresponding moment of the day-ahead load prediction value vector and the historical measured load value , :

[0142] ;

[0143] ;

[0144] At the next moment, the predicted load value to be issued The calculation formula is as follows:

[0145] .

[0146] In this embodiment, the vector of the day-ahead load prediction value is a vector of the load prediction values within the next day, corresponding to the predicted loads at 24 moments respectively. For example, when predicting the load at 10 o'clock, it is necessary to extract the load prediction value corresponding to 10 o'clock in the day-ahead load prediction vector of the current day and the intra-day load prediction value to calculate the credibility together.

[0147] According to the above solution of the present invention, based on the day-ahead load prediction model and the intra-day load prediction model, combined with the measured load data, a prediction and correction method for the end load prediction and adaptive correction system of the ground source heat pump system of the present invention is constructed, as Figure 2 shown, the specific operation steps of the system are as follows:

[0148] Step 1. Update the model coefficients of the day-ahead load prediction model and the day-ahead load prediction deviation correction model, and calculate the vector of the day-ahead load prediction values within one day .

[0149] Step 2. Update the model coefficients of the intra-day load prediction model, and calculate the intra-day load prediction value at the prediction moment .

[0150] Step 3. Calculate the credibility of the prediction value at a historical moment , , and calculate the predicted load value to be issued .

[0151] Step 4. Issue the predicted load value to be issued , and after guiding the actual operation of the system, obtain the actual operation load , and update the historical measured load data set; update the data set of the historical predicted load data according to the day-ahead and intra-day load prediction values; update the data set of the historical predicted meteorological data from the historical outdoor predicted meteorological data.

[0152] Step 5. If the current day has not ended (been predicted), return to Step 2 for intra-day load prediction; if it has ended, judge whether to continue running the next day. If so, obtain the predicted meteorological data for the next day, and then return to Step 1 to run in a loop; otherwise, end the operation.

[0153] According to the above solution of the present invention, by introducing meteorological variables such as ambient temperature and humidity, the present invention constructs a regression model to dynamically correct the prediction results, which helps to improve the prediction accuracy. At the same time, the real-time update and rolling training strategy of the model parameters can ensure that the model adapts to the dynamic changes of the operating environment and enhance the system robustness.

[0154] To achieve closed-loop operation, it is necessary to combine the prediction credibility assessment and the feedback control mechanism to evaluate the model performance in real time and adjust the prediction strategy. This solution can not only improve the load prediction accuracy, reduce energy waste, but also significantly reduce the operating cost and equipment loss, and contribute to the development of green energy.

[0155] According to the above solution of the present invention, by constructing a two-layer load prediction model, the present invention effectively improves the load prediction accuracy and dispatching decision-making level of the cold and heat source system. Its day-ahead prediction model captures the long-term trends and seasonal variations, and the intraday prediction model responds to short-term fluctuations. Combining with the meteorological parameter correction model, it enhances the adaptability of the system to the dynamic changes of the environment. Through the rolling update and feedback control mechanism, the system can automatically adjust the model parameters to ensure the stability and continuous optimization of the prediction results.

[0156] In actual operation, the present invention can significantly reduce energy waste and equipment loss, reduce the operating cost, and improve the reliability and economic benefits of the system operation. In addition, through the dynamic optimization of energy distribution, the system effectively reduces carbon emissions and environmental pollution, and has important social and environmental values. At the same time, the invention has good versatility for various energy systems and can be widely applied to scenarios such as regional energy management, intelligent building control, and industrial process optimization, providing technical support for the comprehensive upgrade of the intelligent energy management system.

[0157] Furthermore, to achieve the above object, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, it implements the method for predicting and adaptively correcting the end load of the ground source heat pump system as described above.

[0158] Furthermore, to achieve the above object, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by the processor, it implements the method for predicting and adaptively correcting the end load of the ground source heat pump system as described above.

[0159] Those of ordinary skill in the art can realize that the modules and algorithm steps described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present invention.

[0160] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described devices and equipment can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0161] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or modules can be electrical, mechanical or other forms.

[0162] The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical modules, that is, they can be located in one place, or can be distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of the embodiments of the present invention.

[0163] In addition, the various functional modules in the embodiments of the present invention can be integrated in one processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0164] When the above functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or a part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method for sending / receiving energy-saving signals in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical discs that can store program codes.

[0165] The above description is only the preferred embodiment of the present application and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, the technical solution formed by mutually replacing the above features with the technical features (but not limited to) having similar functions disclosed in the present application.

[0166] It should be understood that the magnitudes of the sequence numbers of the steps in the content and embodiments of the present invention do not absolutely mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

Claims

1. A method for predicting and adaptively correcting the terminal load of a ground source heat pump system, characterized in that: include: Obtain historical operating load data, historical forecast load data, historical measured load data, and historical outdoor forecast meteorological data; Building a day-ahead load forecasting model based on historical operating load data and historical measured load data, and calculating a day-ahead load forecasting value based on the day-ahead load forecasting model; A day-ahead load forecast deviation correction model is constructed based on historical forecast load data, historical measured load data, and historical outdoor forecast meteorological data, and a day-ahead load forecast deviation correction value is calculated based on the day-ahead load forecast deviation correction model; Obtaining a day-ahead load forecast value vector based on the day-ahead load forecast deviation correction value; A daily load forecasting model is constructed based on historical operating load data, historical forecast load data and historical outdoor forecast meteorological data, and a daily load forecast value is calculated based on the daily load forecasting model, including: The least squares method is used to construct a daily load forecasting model using historical measured load data, historical forecast load data, and historical outdoor forecast meteorological data for at least the past two weeks: ； Where: is the load forecast value at time t; p is the number of hours considered for historical load; is the historical load factor, ; is the load value at the past moment i; q is the number of hours considered for historical residuals; is the historical residual coefficient, ; is the residual at the past i-th moment; is the meteorological parameter coefficient vector; C is the constant term coefficient; To predict outdoor temperature; To predict outdoor relative humidity; The credibility of the intraday load forecast value, the forecast value at the corresponding time in the day-ahead load forecast value vector and the historical measured load data is calculated respectively, and the forecast load delivery value is obtained based on the intraday load forecast value, the forecast value at the corresponding time in the day-ahead load forecast value vector and the credibility calculation, including: Extracting daily load forecast values And the forecast value of the day-ahead load forecast value vector corresponding to the time , calculate the daily load forecast value respectively , the predicted value of the day-ahead load forecast vector at the corresponding time Compared with the historical measured load value Credibility , : ； ； At the next moment, the calculation formula for the forecast load value L is as follows: ； The method of constructing a day-ahead load forecast deviation correction model based on historical forecast load data, historical measured load data, and historical outdoor forecast meteorological data, and calculating a day-ahead load forecast deviation correction value based on the day-ahead load forecast deviation correction model includes: Calculate the residual using historical measured load data and historical predicted load data: ； Where: is the load forecast residual at time t the day before; is the load value at the tth moment; is the load forecast value at the tth moment; A multivariate linear regression approach was used to predict outdoor temperatures using historical outdoor forecast weather data from at least the past two weeks. and predicted outdoor relative humidity , Load forecast residual Construct a day-ahead load forecast deviation correction model: ； After training, the parameters of the day-ahead load forecast deviation correction model are obtained. ; Predicted outdoor temperature at time t+k in the future and predicted outdoor relative humidity , substitute the trained day-ahead load forecast deviation correction model: ； The correction value of the day-ahead load forecast deviation is: ； In the formula, is the basic value of the load forecast at the day before t+k; The day-ahead load forecast value vector obtained based on the day-ahead load forecast deviation correction value is: Based on the day-ahead load forecast deviation correction value , where k is 0-23, then the day-ahead load forecast value vector for the next 24 hours can be obtained: ; 。 2. The method for predicting and adaptively correcting the terminal load of a ground source heat pump system according to claim 1, characterized in that: The method of constructing a day-ahead load forecasting model based on historical operating load data and historical measured load data, and calculating a day-ahead load forecasting value based on the day-ahead load forecasting model includes: Using historical operating load data from at least the past two weeks, the Holt-Winters additive model is constructed as a day-ahead load forecasting model to perform basic day-ahead load forecasting: ； Where: is the basic value of the load forecast at the day before the tth moment; is the load value at the tth moment; is the smoothing value at the tth moment; is the trend value at the tth moment; is the period value at time t; is the cycle length, which is 24 in the day-ahead load forecasting model; is the smoothing coefficient, ; is the trend coefficient, ; is the seasonal coefficient, ; The day-ahead load forecast value is obtained by predicting the future time t+k: .

3. The terminal load prediction and adaptive correction system of the ground source heat pump system is characterized by: include: Data acquisition module, which acquires historical operating load data, historical forecast load data, historical measured load data, and historical outdoor forecast meteorological data; A day-ahead load forecasting model building module is used to build a day-ahead load forecasting model based on historical operating load data and historical measured load data, and to calculate a day-ahead load forecast value based on the day-ahead load forecasting model; A day-ahead load forecast deviation correction model construction module is used to construct a day-ahead load forecast deviation correction model based on historical forecast load data, historical measured load data, and historical outdoor forecast meteorological data, and to calculate a day-ahead load forecast deviation correction value based on the day-ahead load forecast deviation correction model; A day-ahead load forecast value vector acquisition module, which obtains a day-ahead load forecast value vector based on the day-ahead load forecast deviation correction value; The intraday load forecasting model building module builds an intraday load forecasting model based on historical operating load data, historical forecast load data and historical outdoor forecast meteorological data, and calculates the intraday load forecast value based on the intraday load forecasting model, including: The least squares method is used to construct a daily load forecasting model using historical measured load data, historical forecast load data, and historical outdoor forecast meteorological data for at least the past two weeks: ； Where: is the load forecast value at time t; p is the number of hours considered for historical load; is the historical load factor, ; is the load value at the past moment i; q is the number of hours considered for historical residuals; is the historical residual coefficient, ; is the residual at the past i-th moment; is the meteorological parameter coefficient vector; C is the constant term coefficient; To predict outdoor temperature; To predict outdoor relative humidity; The forecast load delivery value acquisition module calculates the credibility of the intraday load forecast value, the forecast value at the corresponding time in the day-ahead load forecast value vector and the historical measured load data, and obtains the forecast load delivery value based on the intraday load forecast value, the forecast value at the corresponding time in the day-ahead load forecast value vector and the credibility, including: Extracting daily load forecast values And the forecast value of the day-ahead load forecast value vector corresponding to the time , calculate the daily load forecast value respectively , the predicted value of the day-ahead load forecast vector at the corresponding time Compared with the historical measured load value Credibility , : ； ； At the next moment, the calculation formula for the forecast load value L is as follows: ； The method of constructing a day-ahead load forecast deviation correction model based on historical forecast load data, historical measured load data, and historical outdoor forecast meteorological data, and calculating a day-ahead load forecast deviation correction value based on the day-ahead load forecast deviation correction model includes: Calculate the residual using historical measured load data and historical predicted load data: ； Where: is the load forecast residual at time t the day before; is the load value at the tth moment; is the load forecast value at the tth moment; A multivariate linear regression approach was used to predict outdoor temperatures using historical outdoor forecast weather data from at least the past two weeks. and predicted outdoor relative humidity , Load forecast residual Construct a day-ahead load forecast deviation correction model: ； After training, the parameters of the day-ahead load forecast deviation correction model are obtained. ; Predicted outdoor temperature at time t+k in the future and predicted outdoor relative humidity , substitute the trained day-ahead load forecast deviation correction model: ； The correction value of the day-ahead load forecast deviation is: ； In the formula, is the basic value of the load forecast at the day before t+k; The day-ahead load forecast value vector obtained based on the day-ahead load forecast deviation correction value is: Based on the day-ahead load forecast deviation correction value , where k is 0-23, then the day-ahead load forecast value vector for the next 24 hours can be obtained: : 。 4. An electronic device, characterized in that The method comprises a processor, a memory and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the method for predicting and adaptively correcting the terminal load of a ground source heat pump system as claimed in claim 1 or 2 is implemented.

5. 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 terminal load prediction and adaptive correction method of the ground source heat pump system as claimed in claim 1 or 2 is implemented.

Images