An Adaptive Agent-based Combined Cooling and Heating Load Management System and Method

Abstract

The present invention discloses a combined cooling and heating power load management system and method based on an adaptive agent, which relates to the technical field of combined cooling and heating management. It includes a combined cooling and heating management module, a prediction data acquisition module, a user-side load analysis module, and a parameter adjustment management module. The combined cooling and heating management module is used to control an air source heat pump device to achieve the combined cooling and heating function and predict the required load at the user side, and adjust and manage the parameters of the air source heat pump device. The prediction data acquisition module is used to collect the water supply temperature data and the historical prediction time data. The user-side load analysis module is used to analyze the historical water supply temperature change time data and plan the load prediction time. The parameter adjustment management module is used to perform load prediction according to the planned time and adjust and manage the parameters of the heat pump device, and uses the adaptive learning method of the agent to perform load prediction and heat pump parameter adjustment, and plan the prediction time, reducing the unnecessary data processing workload while ensuring the energy supply and demand.

Description

Technical Field

[0001] The present invention relates to the technical field of combined cooling and heating supply management, and specifically to a combined cooling and heating supply load management system and method based on an adaptive agent. Background Art

[0002] An air source heat pump is an energy-saving device that uses high-level energy to make heat flow from a low-level heat source, air, to a high-level heat source. It can convert low-level heat energy that cannot be directly utilized into high-level heat energy that can be utilized, thereby achieving the purpose of saving part of the high-level energy. In traditional energy management systems, the operating efficiency of air source heat pumps is often limited by the delay in responding to environmental variables and non-optimal energy allocation. Under changing climate conditions, heat pumps often have difficulty adjusting their operating states in a timely manner to adapt to environmental changes, affecting the overall energy efficiency and heat pump performance. By enabling an agent to adaptively learn to predict loads and timely adjust heat pump parameters, it is possible to more precisely match energy supply and demand, thereby reducing energy waste without sacrificing user comfort and improving the overall energy efficiency of the heat pump;

[0003] However, in the prior art, the method of collecting environmental data in real time and predicting loads in real time is generally used to predict the loads required by the user side. However, it is ignored that the loads required by the user side may remain unchanged for a long time. If the prediction results before and after are the same, in this case, there is no need to adjust the heat pump parameters. For this situation, the prior art does not set an appropriate time for data prediction, and it is impossible to reduce unnecessary data processing workload while ensuring energy supply and demand, reducing the necessity and significance of system prediction and data processing. Summary of the Invention

[0004] The purpose of the present invention is to provide a combined cooling and heating supply load management system and method based on an adaptive agent to solve the problems raised in the prior art.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A combined cooling and heating supply load management system based on an adaptive agent, including a combined cooling and heating supply management module, a prediction data collection module, a user-side load analysis module, and a parameter adjustment management module;

[0006] The combined cooling and heating supply management module is used to control the air source heat pump device to achieve the combined cooling and heating supply function and predict the loads required by the user side, and adjust and manage the parameters of the air source heat pump device;

[0007] The prediction data collection module is used to collect the water supply temperature data and historical prediction time data obtained by default prediction in the past;

[0008] The user-side load analysis module is used to analyze the historical water supply temperature change time data and plan and adjust the load prediction time;

[0009] The parameter adjustment management module is used to perform load forecasting according to the planned time and manage the parameter adjustment of the air source heat pump equipment. The parameter refers to the load required by the user side, that is, the water supply temperature of the air source heat pump equipment.

[0010] Preferably, the combined cooling and heating supply management module includes an air source heat pump control unit, a user-side load forecasting unit, and a water supply temperature adjustment unit;

[0011] The air source heat pump control unit is used to control the air source heat pump equipment to generate hot water or cold water required by the user, and transport the hot water or cold water generated by the heat pump equipment to the user side through the water supply pipeline. After the user uses the hot water, the cooled water returns to the heat pump equipment through the return water pipeline to be reheated. This process reflects the combined cooling and heating supply function of the air source heat pump equipment;

[0012] The user-side load forecasting unit is used to sense environmental data through the self-adaptive learning method of the intelligent agent, establish a user-side load forecasting model using the deep learning algorithm, input the environmental data into the model to predict the load required by the user side. The intelligent agent refers to an intelligent entity, which is a three-dimensional perception, global collaboration, accurate judgment, continuous evolution, and open intelligent system built based on the cloud and with AI as the core. The intelligent agent itself has the ability of self-adaptive learning. The deep learning algorithm refers to the Actor-Critic algorithm. The Actor-Critic algorithm is a deep reinforcement learning algorithm that combines behavior learning and evaluation feedback. The environmental data includes outdoor temperature, indoor temperature, return water temperature, building attribute data, and heating area data. The load required by the user side refers to the water supply temperature;

[0013] The water supply temperature adjustment unit is used to adjust the water supply temperature of the air source heat pump equipment to the predicted water supply temperature.

[0014] Preferably, the prediction data acquisition module includes a water supply temperature data acquisition unit and a data prediction time acquisition unit;

[0015] The water supply temperature data acquisition unit is used to collect the water supply temperature data predicted every day when the water supply temperature was previously predicted in a real-time prediction manner;

[0016] The data prediction time acquisition unit is used to acquire the time corresponding to the predicted water supply temperature.

[0017] Preferably, the user-side load analysis module includes a temperature change time analysis unit and a collection and prediction time planning unit;

[0018] The temperature change time analysis unit is used to count the time when the predicted water supply temperature changes every day in the past, and obtain the default initial time for starting load prediction every day in the past as T. The time interval between the time when the water supply temperature first changes after time T and T is recorded as the initial water supply temperature change time interval. The initial water supply temperature change time interval data of each day in the past are formed into a historical first data set. Analyze the time interval data in the historical first data set to predict the initial water supply temperature change time of the latest day. After predicting the initial water supply temperature change time of the latest day, retrieve the water supply temperature change time data of several days when the initial water supply temperature change time is the same as the predicted initial water supply temperature change time of the latest day, and obtain the time when the water supply temperature first changes after the corresponding initial water supply temperature change time. The time interval between the obtained time and the corresponding initial water supply temperature change time is recorded as the secondary water supply temperature change time interval. The secondary water supply temperature change time interval data of the corresponding several days are formed into a historical second data set. Analyze the historical second data set and predict the secondary water supply temperature change time of the latest day, and so on to predict the change time of the water supply temperature of the latest day several times;

[0019] The acquisition and prediction time planning unit is used to plan the load prediction time and adjust the default real-time prediction method: sense the environmental data at the change times of the water supply temperature of the latest day predicted several times, and predict the load required by the user side, that is, the water supply temperature, and adjust the water supply temperature of the air source heat pump device to the water supply temperature predicted at the corresponding time.

[0020] Preferably, the parameter adjustment management module includes a usage time sensing unit, an adjustment time anomaly prediction unit, and a water temperature adjustment time adjustment unit;

[0021] The usage time sensing unit is used to sense the time period when the user uses hot water on the latest day. A water level sensor is installed in the water tank of the air source heat pump device. The water level is monitored through the water level sensor. When the water level drops, it is judged that the user is using hot water. When the water level stops dropping, it is judged that the user stops using hot water, so as to be able to sense the time period when the user uses hot water;

[0022] The adjustment time anomaly prediction unit is used to judge whether the planned load prediction time falls within the sensed time period: if the planned prediction time is within the sensed time period, the corresponding prediction time is marked as an anomaly; otherwise, no anomaly marking is performed;

[0023] The water temperature adjustment time adjustment unit is used to plan the water supply temperature adjustment time: if none of the planned load prediction times are marked, the water supply temperature of the air source heat pump device is adjusted to the predicted water supply temperature immediately after the water supply temperature is predicted at the corresponding prediction time; if there is an abnormally marked prediction time within the sensed time period, the water supply temperature is adjusted to the water supply temperature predicted at the abnormally marked prediction time after it is sensed that the user stops using hot water; if there are more than one abnormally marked prediction times within the sensed time period, the water supply temperature is adjusted to the water supply temperature predicted at the latest prediction time among several abnormally marked prediction times after it is sensed that the user stops using hot water.

[0024] A combined cooling and heating load management method based on an adaptive agent includes the following steps:

[0025] S100: Control the air source heat pump device to realize the combined cooling and heating function and conduct load prediction required by the user side, and adjust and manage the parameters of the air source heat pump device;

[0026] S200: Collect the water supply temperature data and historical prediction time data predicted by the default method in the past.

[0027] S300: Analyze the historical water supply temperature change time data, plan and adjust the load prediction time;

[0028] S400: Conduct load prediction according to the planned time and adjust and manage the parameters of the air source heat pump device.

[0029] Preferably, in step S100: Control the air source heat pump device to generate hot water or cold water required by the user, transport the hot water or cold water generated by the heat pump device to the user side through the water supply pipeline. After the user uses hot water, the cooled water returns to the heat pump device through the return water pipeline to be reheated. Perceive the environmental data in the way of intelligent agent adaptive learning, establish a user side load prediction model using the deep learning algorithm, input the environmental data into the model to predict the load required by the user side, and adjust the water supply temperature of the air source heat pump device to the predicted water supply temperature.

[0030] Preferably, in step S200: Collect the water supply temperature data predicted every day when predicting the water supply temperature by the default real-time prediction method in the past, and obtain the time when the corresponding water supply temperature is predicted.

[0031] Preferably, in step S300: retrieve the daily predicted water supply temperature data when the real-time prediction method was previously used by default to predict the water supply temperature, count the time when the predicted water supply temperature changed daily in the past, obtain the default initial time T for load prediction every day in the past, record the time interval between the time when the water supply temperature first changed after time T and T as the initial water supply temperature change time interval, and form a historical first data set with the initial water supply temperature change time interval data for each day in the past, obtaining the historical first data set as {t1, t2,..., t k}, where k represents the number of data items in the historical first data set, and t k represents the initial water supply temperature change time interval for the kth day in the past, and establish a first time interval prediction model:

[0032] R1 = λ * t k + (1 - λ) * ζ k ;

[0033] Predict that the time interval between the initial water supply temperature change time of the latest day and T is R1, and the initial water supply temperature change time of the latest day is A1: A1 = T + R1, where * represents the multiplication sign, λ represents the smoothing coefficient, 0 < λ < 1, and ζ k represents the exponential smoothing value of the initial water supply temperature change time interval for the kth day in the past, and solve ζ k as follows: Calculate the exponential smoothing value ζ1 of the initial water supply temperature change time interval for the 1st day in the past according to ζ1 = λ * t1 + (1 - λ) * [(t1 + t2 + t3) / 3], calculate the exponential smoothing value ζ2 of the initial water supply temperature change time interval for the 2nd day in the past according to ζ2 = λ * t1 + (1 - λ) * ζ1, calculate the exponential smoothing value ζ3 of the initial water supply temperature change time interval for the 2nd day in the past according to ζ3 = λ * t2 + (1 - λ) * ζ2, and so on to gradually solve for ζ k , after predicting R1, retrieve that there are m days when the initial water supply temperature change time is the same as T + R1, retrieve the water supply temperature change data for the corresponding m days, obtain the time when the water supply temperature first changes after T + R1, record the time interval between the obtained time and T + R1 as the secondary water supply temperature change time interval, and form a historical second data set with the secondary water supply temperature change time interval data for the corresponding m days, obtaining the historical second data set as {U1, U2,..., U m}, and establish a second time interval prediction model:

[0034] R2 = λ * U m + (1 - λ) * η m ;

[0035] The time interval between the time of the second change in the water supply temperature predicted for the latest day and T+R1 is R2, and the time of the second change in the water supply temperature for the latest day is A2: A2 = T + R1 + R2, η m represents the exponentially smoothed value of the time interval of the second change in the water supply temperature corresponding to the m-th day, η m The solution method of k is the same as that of ζ. By analogy, the data set of the change times of the water supply temperature for the latest day predicted is {A1, A2,..., A n}, where A n represents the time of the n-th change in the water supply temperature predicted for the latest day. According to the time-aware environmental data in the data set {A1, A2,..., A n}, predict the load required by the user side, that is, the water supply temperature, and adjust the water supply temperature of the air source heat pump device to the water supply temperature predicted at the corresponding time;

[0036] To solve the problem that the real-time prediction method cannot reduce the unnecessary data processing workload when the load required by the user side may remain unchanged for a long time, collect the water supply temperature data predicted by the previous real-time prediction method and the time data of the change in the water supply temperature through big data technology. Considering that the initial time for the system to start collecting environmental data for load prediction every day is the same, taking the default initial time as the benchmark, analyze the change of the time interval between the predicted change time of the water supply temperature in the historical data and the initial time, establish a first time interval prediction model through the exponential smoothing algorithm, and predict the time when the water supply temperature of the heat pump device changes for the first time on the latest day. Considering that the first change in the water supply temperature every day may be different, taking the predicted initial change time of the water supply temperature on the latest day as the benchmark, screen out the data in the historical data where the initial change time of the water temperature is the same as the predicted initial change time, and continue to predict the time when the water supply temperature changes for the second time on the latest day based on the selected data. By analogy, screen out the historical data with the same time as the previous prediction time as a reference, and continue to predict the time of subsequent several changes in the water supply temperature on the latest day, effectively improving the accuracy of the data prediction result. According to the predicted time, sense the environmental data and predict the load required by the user side, that is, the water supply temperature, realize the dynamic adjustment of the load prediction time, then adjust the water supply temperature of the heat pump according to the load prediction result, analyze the historical water supply temperature change law, and perform load prediction when the water supply temperature may change, reducing the unnecessary data processing workload while ensuring the energy supply and demand, and improving the necessity and significance of the system's prediction and data processing.

[0037] Preferably, in step S400: It is sensed that any hot water usage time period of the user on the latest day is [H1, H2], where H1 represents the start time of the corresponding time period and H2 represents the end time of the corresponding time period. It is judged whether the load prediction time in the data set {A1, A2,..., A n} falls within the time period [H1, H2]. It is statistically found that the times A1 and A2 fall within the time period [H1, H2], that is, the time H1 is before the time A1, the time A1 is before the time A2, and the time A2 is before the time H2. The times A1 and A2 are marked as abnormal, and at the moment of H2, the water supply temperature of the heat pump device is adjusted to the water supply temperature predicted at the time A2;

[0038] After the load prediction operation is completed, further considering that the load prediction time may be within the user's hot water usage time period, at this time, immediately after predicting the load required by the user side, adjusting the water supply temperature of the heat pump device may cause an impact or even discomfort to the user's hot water usage experience. When it is judged that this situation occurs, instead of immediately adjusting the water supply temperature of the heat pump after predicting the water supply temperature, the water supply temperature of the heat pump is adjusted to an appropriate temperature when the user stops using hot water, effectively avoiding the situation that the sudden change in temperature during hot water usage causes an impact or even discomfort to the user's hot water usage experience.

[0039] Compared with the prior art, the beneficial effects of the present invention are:

[0040] The present invention uses the method of adaptive learning of agents for load forecasting and parameter adjustment of air source heat pump equipment, effectively ensuring the energy supply and demand; collects the water supply temperature data predicted by the previous real-time prediction method and the time data of the change in water supply temperature through big data technology. Considering that the initial time for collecting ambient data for load forecasting set by the system by default is the same every day, taking the default initial time as the benchmark, analyzes the change in the time interval between the predicted water supply temperature change time and the initial time in historical data, and establishes a first time interval prediction model through the exponential smoothing algorithm to predict the time when the water supply temperature of the heat pump equipment changes for the first time on the latest day. Considering that the time when the water supply temperature changes for the first time every day in the past may be different, taking the predicted initial change time of the water supply temperature on the latest day as the benchmark, screens out the data with the same initial water temperature change time as the predicted initial change time from historical data, and continues to predict the time when the water supply temperature changes for the second time on the latest day based on the screened data, and so on, screens out the historical data with the same time as the previous prediction time as a reference, and continues to predict the time of subsequent several changes in the water supply temperature on the latest day, effectively improving the accuracy of the data prediction result. Perceives the ambient data according to the predicted time and predicts the load required by the user side, that is, the water supply temperature, realizes the dynamic adjustment of the load forecasting time, and then adjusts the water supply temperature of the heat pump according to the load forecasting result, analyzes the historical water supply temperature change law, and performs load forecasting when the water supply temperature may change, reducing the unnecessary data processing workload while ensuring the energy supply and demand, and improving the necessity and significance of the system for predicting and processing data;

[0041] After the load forecasting operation is completed, further considering that the load forecasting time may be within the user's hot water usage period, at this time, after predicting the load required by the user side, immediately adjusting the water supply temperature of the heat pump equipment, the sudden change in temperature may affect or even discomfort the user's hot water usage experience. When it is judged that this situation occurs, after predicting the water supply temperature, do not immediately adjust the water supply temperature of the heat pump, but adjust the water supply temperature of the heat pump to an appropriate temperature when the user stops using hot water, effectively avoiding the situation that the sudden change in temperature during the hot water usage process affects or even discomfort the user's hot water usage experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic structural diagram of a combined cooling and heating load management system based on an adaptive agent of the present invention;

[0043] Figure 2 It is a schematic flow diagram of a combined cooling and heating load management method based on an adaptive agent of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0044] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0045] Embodiment 1:

[0046] As Figure 1 shown, this embodiment provides a combined cooling and heating load management system based on an adaptive agent. The system includes: a combined cooling and heating management module, a prediction data acquisition module, a user-side load analysis module, and a parameter adjustment management module;

[0047] The output end of the combined cooling and heating management module is connected to the input end of the prediction data acquisition module, the output end of the prediction data acquisition module is connected to the input end of the user-side load analysis module, and the output end of the user-side load analysis module is connected to the input end of the parameter adjustment management module;

[0048] The combined cooling and heating management module is used to control the air source heat pump equipment to achieve the combined cooling and heating function and perform the load prediction required by the user side, and adjust and manage the parameters of the air source heat pump equipment;

[0049] The prediction data acquisition module is used to collect the water supply temperature data and historical prediction time data obtained by default prediction in the past;

[0050] The user-side load analysis module is used to analyze the historical water supply temperature change time data and plan and adjust the load prediction time;

[0051] The parameter adjustment management module is used to perform load prediction according to the planned time and adjust and manage the parameters of the air source heat pump equipment. The parameter refers to the load required by the user side, that is, the water supply temperature of the air source heat pump equipment.

[0052] The combined cooling and heating management module includes an air source heat pump control unit, a user-side load prediction unit, and a water supply temperature adjustment unit;

[0053] The output end of the air source heat pump control unit is connected to the input end of the user-side load prediction unit, and the output end of the user-side load prediction unit is connected to the input end of the water supply temperature adjustment unit;

[0054] The air source heat pump control unit is used to control the air source heat pump equipment to generate hot water or cold water required by the user, and transport the hot water or cold water generated by the heat pump equipment to the user side through the water supply pipeline. After the user uses the hot water, the cooled water returns to the heat pump equipment through the return water pipeline to be reheated. This process reflects the combined cooling and heating function of the air source heat pump equipment;

[0055] The client load prediction unit is used to sense environmental data through the self-adaptive learning of agents, establish a client load prediction model using deep learning algorithms, input the environmental data into the model to predict the load required by the client. An agent refers to an intelligent entity, which is a three-dimensional perception, global collaboration, precise judgment, continuous evolution, and open intelligent system built based on the cloud and centered on AI. The agent itself has the ability of self-adaptive learning. The deep learning algorithm refers to the Actor-Critic algorithm, which is a deep reinforcement learning algorithm that combines behavior learning and evaluation feedback. The environmental data includes outdoor temperature, indoor temperature, return water temperature, building attribute data, and heating area data. The load required by the client refers to the supply water temperature;

[0056] The supply water temperature regulation unit is used to adjust the supply water temperature of the air source heat pump device to the predicted supply water temperature.

[0057] The prediction data acquisition module includes a supply water temperature data acquisition unit and a data prediction time acquisition unit;

[0058] The input end of the supply water temperature data acquisition unit is connected to the output end of the supply water temperature regulation unit, and the output end of the supply water temperature data acquisition unit is connected to the input end of the data prediction time acquisition unit;

[0059] The supply water temperature data acquisition unit is used to collect the supply water temperature data predicted every day when the supply water temperature was previously predicted by the default real-time prediction method;

[0060] The data prediction time acquisition unit is used to acquire the time corresponding to the predicted supply water temperature.

[0061] The client load analysis module includes a temperature change time analysis unit and a collection prediction time planning unit;

[0062] The input end of the temperature change time analysis unit is connected to the output end of the data prediction time acquisition unit, and the output end of the temperature change time analysis unit is connected to the input end of the collection prediction time planning unit;

[0063] The temperature change time analysis unit is used to count the time when the predicted water supply temperature changes every day in the past, and obtain the default initial time T for starting load prediction every day in the past. The time interval between the time when the water supply temperature first changes after time T and T is recorded as the initial water supply temperature change time interval. The initial water supply temperature change time interval data of each day in the past are formed into a historical first data set. Analyze the time interval data in the historical first data set to predict the initial water supply temperature change time of the latest day. After predicting the initial water supply temperature change time of the latest day, retrieve the water supply temperature change time data of several days when the initial water supply temperature change time is the same as the predicted initial water supply temperature change time of the latest day, and obtain the time when the water supply temperature first changes after the corresponding initial water supply temperature change time. The time interval between the obtained time and the corresponding initial water supply temperature change time is recorded as the secondary water supply temperature change time interval. The secondary water supply temperature change time interval data of the corresponding several days are formed into a historical second data set. Analyze the historical second data set and predict the secondary water supply temperature change time of the latest day, and so on to predict the change time of the water supply temperature of the latest day for several times;

[0064] The acquisition prediction time planning unit is used to plan the load prediction time and adjust the default real-time prediction method: sense the environmental data at the change times of the water supply temperature of the latest day predicted, and predict the load required by the user side, that is, the water supply temperature, and adjust the water supply temperature of the air source heat pump device to the water supply temperature predicted at the corresponding time.

[0065] The parameter adjustment management module includes a usage time sensing unit, an adjustment time anomaly prediction unit, and a water temperature adjustment time adjustment unit;

[0066] The output end of the usage time sensing unit and the acquisition prediction time planning unit is connected to the input end of the adjustment time anomaly prediction unit, and the output end of the adjustment time anomaly prediction unit is connected to the input end of the water temperature adjustment time adjustment unit;

[0067] The usage time sensing unit is used to sense the time period when the user uses hot water on the latest day. A water level sensor is installed in the water tank of the air source heat pump device. The water level is monitored through the water level sensor. When the water level drops, it is judged that the user is using hot water. When the water level stops dropping, it is judged that the user stops using hot water, so as to be able to sense the time period when the user uses hot water;

[0068] The adjustment time anomaly prediction unit is used to judge whether the planned load prediction time falls within the sensed time period: if the planned prediction time is within the sensed time period, the corresponding prediction time is marked as abnormal; otherwise, it is not marked as abnormal;

[0069] The water temperature adjustment time adjustment unit is used to plan the water supply temperature adjustment time: If none of the planned load prediction times are marked, the water supply temperature of the air source heat pump device is adjusted to the predicted water supply temperature immediately after the water supply temperature is predicted at the corresponding prediction time; If there is an abnormally marked prediction time within the sensed time period, the water supply temperature is adjusted to the water supply temperature predicted at the abnormally marked prediction time after sensing that the user stops using hot water; If there are more than one abnormally marked prediction times within the sensed time period, the water supply temperature is adjusted to the water supply temperature predicted at the latest prediction time among several abnormally marked prediction times after sensing that the user stops using hot water.

[0070] Embodiment 2:

[0071] As Figure 2 shown, this embodiment provides a combined cooling and heating load management method based on an adaptive agent, which is implemented based on the combined cooling and heating load management system in the embodiment, and specifically includes the following steps:

[0072] S100: Control the air source heat pump device to realize the combined cooling and heating function and predict the load required by the user side, adjust and manage the parameters of the air source heat pump device, control the air source heat pump device to generate the hot water or cold water required by the user, transport the hot water or cold water generated by the heat pump device to the user side through the water supply pipeline. After the user uses the hot water, the cooled water returns to the heat pump device through the return water pipeline to be reheated. Perceive the environmental data through the self-adaptive learning method of the agent, establish a user-side load prediction model using the deep learning algorithm, input the environmental data into the model to predict the load required by the user side, and adjust the water supply temperature of the air source heat pump device to the predicted water supply temperature;

[0073] S200: Collect the water supply temperature data and historical prediction time data predicted by the default method in the past, collect the water supply temperature data predicted every day when the water supply temperature was predicted by the real-time prediction method by default in the past, and obtain the time when the corresponding water supply temperature was predicted;

[0074] S300: Analyze the historical water supply temperature change time data, plan and adjust the load prediction time, retrieve the water supply temperature data predicted every day when the water supply temperature was predicted by the real-time prediction method by default in the past, count the time when the water supply temperature changed every day in the past, obtain the default initial time T for starting load prediction every day in the past, record the time interval between the time when the water supply temperature first changed after the T moment and T as the water supply temperature initial change time interval, and form a historical first data set with the water supply temperature initial change time interval data of each day in the past, and obtain the historical first data set as {t1, t2,..., t k}, where k represents the number of data items in the historical first data set, and tk Denote the initial change time interval of the water supply temperature on the k-th previous day, and establish the first time interval prediction model:

[0075] R1 = λ * t k +(1 - λ) * ζ k ;

[0076] The predicted time interval between the initial change time of the water supply temperature on the latest day and T is R1, and the initial change time of the water supply temperature on the latest day is A1: A1 = T + R1, where λ represents the smoothing coefficient, 0 < λ < 1, and ζ k represents the exponential smoothing value of the initial change time interval of the water supply temperature on the k-th previous day. ζ is solved by the following method k : Calculate the exponential smoothing value ζ1 of the initial change time interval of the water supply temperature on the 1st previous day according to ζ1 = λ * t1+(1 - λ) * [(t1 + t2 + t3) / 3]. Calculate the exponential smoothing value ζ2 of the initial change time interval of the water supply temperature on the 2nd previous day according to ζ2 = λ * t1+(1 - λ) * ζ1. Calculate the exponential smoothing value ζ3 of the initial change time interval of the water supply temperature on the 2nd previous day according to ζ3 = λ * t2+(1 - λ) * ζ2, and so on to gradually solve for ζ k , after predicting R1, it is found that there are m days when the initial change time of the water supply temperature is the same as T + R1. Retrieve the water supply temperature change data for the corresponding m days, and obtain the time when the water supply temperature first changes after T + R1. Denote the time interval between the obtained time and T + R1 as the secondary change time interval of the water supply temperature. Form the historical second data set with the secondary change time interval data of the corresponding m days of the water supply temperature, and obtain the historical second data set as {U1, U2,..., U m}, and establish the second time interval prediction model:

[0077] R2 = λ * U m +(1 - λ) * η m ;

[0078] The predicted time interval between the secondary change time of the water supply temperature on the latest day and T + R1 is R2, and the secondary change time of the water supply temperature on the latest day is A2: A2 = T + R1 + R2, and η m represents the exponential smoothing value of the secondary change time interval of the water supply temperature corresponding to the m-th day. The solution method of η m is the same as that of ζ k , and so on. The predicted change time data set of the water supply temperature on the latest day is {A1, A2,..., A n}, and A n represents the time of the n-th change of the water supply temperature on the latest day predicted. According to the data set {A1, A2,..., An The time-aware environmental data in} are sensed and the required load at the user side, i.e., the water supply temperature, is predicted, and the water supply temperature of the air source heat pump device is adjusted to the water supply temperature predicted for the corresponding time.

[0079] S400: Perform load prediction according to the planned time and manage the parameter adjustment of the air source heat pump device. Sense the time period when the user uses hot water on the latest day, and determine whether the planned load prediction time falls within the sensed time period: If the planned prediction time is within the sensed time period, mark the corresponding prediction time as abnormal; otherwise, do not mark it as abnormal and perform the water supply temperature adjustment time planning: If none of the planned load prediction times are marked, immediately adjust the water supply temperature of the air source heat pump device to the predicted water supply temperature after predicting the water supply temperature at the corresponding prediction time; If there is one abnormally marked prediction time within the sensed time period, adjust the water supply temperature to the water supply temperature predicted at the abnormally marked prediction time after sensing that the user stops using hot water; If there are more than one abnormally marked prediction times within the sensed time period, adjust the water supply temperature to the water supply temperature predicted at the latest prediction time among several abnormally marked prediction times after sensing that the user stops using hot water.

[0080] For example: It is sensed that any time period when the user uses hot water on the latest day is [H1, H2], where H1 represents the start time of the corresponding time period and H2 represents the end time of the corresponding time period. Determine whether the load prediction times in the data set {A1, A2,..., A n} fall within the time period [H1, H2]. It is statistically found that the times A1 and A2 fall within the time period [H1, H2], that is, the time H1 is before the time A1, and the time A1 is before the time A2, and the time A2 is before the time H2. Mark the times A1 and A2 as abnormal, and adjust the water supply temperature of the heat pump device to the water supply temperature predicted at the time A2 at the time H2.

[0081] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, in any regard, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any reference signs in the claims should not be regarded as limiting the claimed rights.

Claims

1. An adaptive agent-based combined cooling and heating power load management system, characterized in that: It includes a combined cooling and heating supply management module, a prediction data acquisition module, a user-side load analysis module, and a parameter adjustment management module; The combined cooling and heating supply management module is used to control the air source heat pump equipment to achieve the combined cooling and heating supply function and conduct load prediction required by the user side, and adjust and manage the parameters of the air source heat pump equipment; The prediction data acquisition module is used to collect the water supply temperature data and historical prediction time data predicted by the previous real-time prediction method; The user-side load analysis module is used to analyze the historical water supply temperature change time data and plan and adjust the load prediction time; The parameter adjustment management module is used to conduct load prediction according to the planned time and adjust and manage the parameters of the air source heat pump equipment.

2. The combined cooling and heating power load management system based on an adaptive agent according to claim 1, wherein: The combined cooling and heating supply management module includes an air source heat pump control unit, a user-side load prediction unit, and a water supply temperature adjustment unit; The air source heat pump control unit is used to control the air source heat pump equipment to generate hot water or cold water required by the user, transport the hot water or cold water generated by the heat pump equipment to the user side through the water supply pipeline, and after the user uses the hot water, the cooled water returns to the heat pump equipment through the return water pipeline to be reheated; The user-side load prediction unit is used to sense the environmental data through the way of intelligent agent adaptive learning, establish a user-side load prediction model by using the deep learning algorithm, input the environmental data into the model to predict the load required by the user side, and the load required by the user side refers to the water supply temperature; The water supply temperature adjustment unit is used to adjust the water supply temperature of the air source heat pump equipment to the predicted water supply temperature.

3. The combined cooling and heating power load management system based on an adaptive agent according to claim 2, wherein: The prediction data acquisition module includes a water supply temperature data acquisition unit and a data prediction time acquisition unit; The water supply temperature data acquisition unit is used to collect the water supply temperature data predicted every day when the water supply temperature is predicted by the previous default real-time prediction method; The data prediction time acquisition unit is used to acquire the time corresponding to the predicted water supply temperature.

4. The combined cooling and heating power load management system based on an adaptive agent according to claim 3, characterized in that: The user-side load analysis module includes a temperature change time analysis unit and a collection prediction time planning unit; The temperature change time analysis unit is used to count the time when the predicted water supply temperature changes every day in the past, obtain the default initial time T for starting load prediction every day in the past, record the time interval between the time when the water supply temperature first changes after time T and T as the initial water supply temperature change time interval, form a historical first data set with the initial water supply temperature change time interval data of each day in the past, analyze the time interval data in the historical first data set, predict the initial water supply temperature change time of the latest day, after predicting the initial water supply temperature change time of the latest day, retrieve the water supply temperature change time data of several days when the initial water supply temperature change time is the same as the predicted initial water supply temperature change time of the latest day, obtain the time when the water supply temperature first changes after the corresponding initial water supply temperature change time, record the time interval between the obtained time and the corresponding initial water supply temperature change time as the secondary water supply temperature change time interval, form a historical second data set with the secondary water supply temperature change time interval data of the corresponding several days, analyze the historical second data set and predict the secondary water supply temperature change time of the latest day, and so on to predict the change time of the water supply temperature of the latest day for several times; The acquisition and prediction time planning unit is used to plan the load prediction time and adjust the default real-time prediction method: sense the environmental data at the change time of the water supply temperature of the latest day predicted, and predict the load required by the user side, that is, the water supply temperature, and adjust the water supply temperature of the air source heat pump device to the water supply temperature predicted at the corresponding time.

5. The combined cooling and heating power load management system based on an adaptive agent according to claim 4, characterized in that: The parameter adjustment management module includes a usage time sensing unit, an adjustment time abnormality prediction unit, and a water temperature adjustment time adjustment unit; The usage time sensing unit is used to sense the time period when the user uses hot water on the latest day; The adjustment time abnormality prediction unit is used to judge whether the planned load prediction time falls within the sensed time period: if the planned prediction time is within the sensed time period, mark the corresponding prediction time as abnormal; Otherwise, no abnormal marking is performed; The water temperature adjustment time adjustment unit is used to plan the water supply temperature adjustment time: if none of the planned load prediction times are marked, immediately adjust the water supply temperature of the air source heat pump device to the predicted water supply temperature after predicting the water supply temperature at the corresponding prediction time; if there is one abnormally marked prediction time within the sensed time period, adjust the water supply temperature to the water supply temperature predicted at the abnormally marked prediction time after sensing that the user stops using hot water; If there are more than one abnormally marked prediction times within the sensed time period, adjust the water supply temperature to the water supply temperature predicted at the latest prediction time among the several abnormally marked prediction times after sensing that the user stops using hot water.

6. A combined cooling and heating power load management method based on adaptive agents, characterized in that: Including the following steps: S100: Control the air source heat pump device to realize the combined cooling and heating function and perform load prediction required by the user side, and adjust and manage the parameters of the air source heat pump device; S200: Collect the water supply temperature data and historical prediction time data predicted by the real-time prediction method in the past; S300: Analyze the historical water supply temperature change time data, plan and adjust the load prediction time; S400: Conduct load prediction according to the planned time and perform parameter adjustment management for the air source heat pump equipment.

7. A combined cooling and heating power load management method based on an adaptive agent according to claim 6, characterized in that: In step S100: Control the air source heat pump equipment to generate the hot water or cold water required by the user, transport the hot water or cold water generated by the heat pump equipment to the user end through the water supply pipeline. After the user uses the hot water, the cooled water returns to the heat pump equipment through the return water pipeline for reheating. Sense the environmental data in the way of intelligent agent adaptive learning, establish a load prediction model for the user end by using the deep learning algorithm, input the environmental data into the model to predict the load required by the user end, and adjust the water supply temperature of the air source heat pump equipment to the predicted water supply temperature.

8. A method for combined cooling and heating power load management based on an adaptive agent according to claim 7, characterized in that: In step S200: Collect the water supply temperature data predicted every day when predicting the water supply temperature in the past by using the real-time prediction method by default, and obtain the time corresponding to the predicted water supply temperature.

9. The method for combined cooling and heating power load management based on an adaptive agent according to claim 8, wherein: In step S300: Retrieve the water supply temperature data predicted every day when the real-time prediction method was previously used by default to predict the water supply temperature, count the time when the predicted water supply temperature changed every day in the past, and obtain the default initial time T for starting the load prediction every day in the past. Denote the time interval between the time when the water supply temperature first changes after time T and T as the initial water supply temperature change time interval. Compose the initial water supply temperature change time interval data for each day in the past into a historical first data set, and obtain the historical first data set as {t1, t2,..., t k}, where k represents the number of data items in the historical first data set, and t k represents the initial water supply temperature change time interval for the k-th day in the past. Establish a first time interval prediction model: R1 = λ * t k + (1 - λ) * ζ k ; The time interval between the initial change time of the water supply temperature on the latest day obtained by prediction and T is R1, and the initial change time of the water supply temperature on the latest day is A1: A1 = T + R1, where λ represents the smoothing coefficient, 0 < λ < 1, and ζ k represents the exponentially smoothed value of the initial change time interval of the water supply temperature on the kth previous day, and ζ is solved by the following method k : The exponentially smoothed value ζ1 of the initial change time interval of the water supply temperature on the 1st previous day is calculated according to ζ1 = λ * t1 + (1 - λ) * [(t1 + t2 + t3) / 3]. The exponentially smoothed value ζ2 of the initial change time interval of the water supply temperature on the 2nd previous day is calculated according to ζ2 = λ * t1 + (1 - λ) * ζ1. The exponentially smoothed value ζ3 of the initial change time interval of the water supply temperature on the 2nd previous day is calculated according to ζ3 = λ * t2 + (1 - λ) * ζ2, and so on to gradually solve for ζ k , after predicting R1, it is found that there are m days when the initial change time of the water supply temperature is the same as T + R1. The water supply temperature change data for the corresponding m days are retrieved, and the time when the water supply temperature first changes after T + R1 is obtained from them. The time interval between the obtained time and T + R1 is recorded as the secondary change time interval of the water supply temperature. The secondary change time interval data of the corresponding m days of the water supply temperature are formed into a historical second data set, and the historical second data set is obtained as {U1, U2,..., U m}, and a second time interval prediction model is established: R2 = λ * U m +(1 - λ) * η m ; The time interval between the predicted second change time of the water supply temperature on the latest day and T+R1 is R2, and the second change time of the water supply temperature on the latest day is A2: A2 = T+R1+R2, η m represents the exponentially smoothed value of the time interval of the second change of the water supply temperature corresponding to the mth day, η m The solution method of k is the same as that of ζ. By analogy, the predicted change time dataset of the water supply temperature on the latest day is {A1, A2,..., A n}, A n represents the time of the nth change of the water supply temperature on the latest day predicted. According to the time perception environmental data in the dataset {A1, A2,..., A n}, predict the required load at the user side, that is, the water supply temperature, and adjust the water supply temperature of the air source heat pump device to the water supply temperature predicted at the corresponding time.

10. A method for combined cooling and heating power load management based on an adaptive agent according to claim 9, characterized in that: In step S400: It is sensed that the time period of any hot water usage by the user on the latest day is [H1, H2], where H1 represents the start time of the corresponding time period and H2 represents the end time of the corresponding time period. It is judged whether the load prediction times in the data set {A1, A2,..., A n} fall within the time period [H1, H2]. It is statistically found that the times A1 and A2 fall within the time period [H1, H2]. The times A1 and A2 are marked as abnormal, and at the moment of H2, the water supply temperature of the heat pump device is adjusted to the water supply temperature predicted at time A2.

Images