CN112268350B - Air conditioner side load prediction method based on system delay - Google Patents
Air conditioner side load prediction method based on system delay Download PDFInfo
- Publication number
- CN112268350B CN112268350B CN202011139927.5A CN202011139927A CN112268350B CN 112268350 B CN112268350 B CN 112268350B CN 202011139927 A CN202011139927 A CN 202011139927A CN 112268350 B CN112268350 B CN 112268350B
- Authority
- CN
- China
- Prior art keywords
- air conditioner
- prediction
- load
- indoor
- prediction model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
- F24F11/46—Improving electric energy efficiency or saving
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/50—Control or safety arrangements characterised by user interfaces or communication
- F24F11/61—Control or safety arrangements characterised by user interfaces or communication using timers
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
- F24F11/63—Electronic processing
- F24F11/64—Electronic processing using pre-stored data
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/70—Control systems characterised by their outputs; Constructional details thereof
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/88—Electrical aspects, e.g. circuits
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2110/00—Control inputs relating to air properties
- F24F2110/10—Temperature
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2110/00—Control inputs relating to air properties
- F24F2110/10—Temperature
- F24F2110/12—Temperature of the outside air
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2130/00—Control inputs relating to environmental factors not covered by group F24F2110/00
- F24F2130/10—Weather information or forecasts
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Human Computer Interaction (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The invention discloses an air conditioner side load prediction method based on system delay, which considers the influence of indoor and outdoor parameters and air conditioner output load from the past to the future, and directly predicts the load to be output by the current air conditioner by utilizing an advanced artificial neural network model. The invention takes the system delay into full consideration, takes the past indoor and outdoor parameters, the air-conditioning load and the current and future indoor and outdoor parameters as the basis of load prediction, and embodies the change characteristics of the system parameters on the time dimension; the prediction model which best meets the target room can be screened out, and the prediction precision is further improved.
Description
Technical Field
The invention relates to the field of indoor thermal environment dynamic regulation and control, in particular to an air conditioner side load prediction method based on system delay.
Background
In the field of indoor thermal environment control, a passive control method is mostly adopted at present, namely future control parameters are determined according to the past system state, for example, a traditional PID controller is used for controlling the room temperature, although the method can meet the temperature control requirement in a certain range, the high-frequency conversion working condition of an air conditioning system is often needed, and therefore the energy consumption and the abrasion of equipment are increased. If an active control method is adopted, the control parameters are optimized in advance by predicting the future output of the system, so that the operation condition of the air-conditioning system can be improved, and the flexibility of the system is improved.
Prediction of the indoor thermal environment is a key step in achieving active control. The cold (heat) load is an important basis when the air conditioning system is designed and controlled, the traditional load is calculated according to parameters such as outdoor meteorological parameters, maintenance structures, indoor heat sources and the like, and the existing load prediction method is also mostly based on the method. However, the predicted load of the existing method focuses on the time-by-time demand load of the room, neglecting the influence of the air conditioning system and the heat transfer delay inside the room. Therefore, the load to be provided by the actual air conditioning system is not necessarily equal to the indoor demand load predicted by the conventional calculation method, subject to the influence of the past system parameters on the current system state and the influence of the current and past system parameters on the future system state. Some improvements provide demand side room loading by the system ahead of time by calculating the delay time of heat transfer for the system and the room. However, if the delay time is longer than 1 hour, and the demand load prediction of the rooms time by time is still performed, the output load state of the air conditioner side is affected at each control time node, and the output load fluctuation of the air conditioner side is large, and the control is inaccurate.
Disclosure of Invention
Based on the prior art, the invention provides an air conditioner side load prediction method based on system delay, which considers the influence of indoor and outdoor parameters and air conditioner output load from the past to the future in terms of air conditioner side output load, and directly predicts the load to be output by the current air conditioner by utilizing an advanced artificial neural network model.
The invention discloses an air conditioner side load prediction method based on system delay, which comprises the following steps:
step 1, collecting historical data of outdoor meteorological parameters, indoor thermal environment parameters and air conditioner side output loads of a target room;
step 4, establishing a prediction model of the air conditioner output load by using the sample obtained in the step 3: the input variable of the prediction model is t0、t1、t2Corresponding indoor and outdoor parameters and Δ t01Average air conditioner output load of corresponding time interval, output is delta t12Dividing all samples into a training sample set and a testing sample set according to the proportion of the average air conditioner output load of the corresponding time interval, training a prediction model by using the training sample set, detecting by using the testing sample set to obtain the prediction error of the prediction model, and storing the trained prediction model and the corresponding delta t01、Δt12;
step 6, until the detection by the test sample is finished, corresponding to all different delta t01、Δt12The prediction errors of the combined prediction models are compared with the prediction precision of the prediction models of different time scales;
Compared with the prior art, the air conditioner side load prediction method based on the system delay has the following advantages:
(1) the system delay is fully considered, the past indoor and outdoor parameters, the air-conditioning load and the current and future indoor and outdoor parameters are taken as the basis of load prediction, and the change characteristics of the system parameters on the time dimension are embodied;
(2) the problem that the traditional calculation indoor load is not matched with the actual output load of the air conditioner due to system delay is effectively solved;
(3) the influence of different combinations of time intervals from the past to the present and from the present to the future on the prediction result is established and compared, the delay of the system is digitalized, the prediction model which is most consistent with the target room is screened out, and the prediction precision is further improved.
Drawings
Fig. 1 is a flowchart of an air conditioner side load prediction method based on system latency according to the present invention.
Detailed Description
The following detailed description of the embodiments of the present invention is provided in connection with the accompanying drawings.
Fig. 1 is a flowchart of an air conditioner side load prediction method based on system delay according to the present invention. The invention mainly comprises the following steps:
step 1, collecting historical data of target room outdoor meteorological parameters, indoor thermal environment parameters and air conditioner side output loads (unit supply and return water temperature difference multiplied by flow) as much as possible;
step 4, utilizing the sample obtained in the step 3 to constructA prediction model of the output load of the vertical air conditioner: the input variable of the prediction model is t0、t1、t2Corresponding indoor and outdoor parameters and Δ t01Average air conditioner output load of corresponding time interval, output is delta t12Dividing all samples into a training sample set and a testing sample set according to the proportion of the average air conditioner output load of the corresponding time interval, training a prediction model by using the training sample set, detecting by using the testing sample set to obtain the prediction error of the prediction model, and storing the trained prediction model and the corresponding delta t01、Δt12;
The prediction model expression, i.e., y ═ f (x), where y denotes the predicted load and x denotes: indoor and outdoor parameters and load at the current moment, and distance delta t from the current moment01Indoor and outdoor parameters and load before the moment, current delta t12Indoor and outdoor parameters after the moment. The prediction model can be established by adopting a regression analysis method or a machine learning algorithm (such as a neural network, a support vector machine, a random forest and the like), and the specific method depends on the accuracy of the model to be achieved and the applicable occasion;
Δt01、Δt12the control method is mainly selected according to the time lag of the system and the expected control precision of the air conditioning system. For example, parameters such as water supply/air supply temperature and flow rate of an air conditioning system change frequently and are unstable, the time interval is shorter in the case, but if the regulation is performed too frequently, such as one minute, the energy consumption of the system operation is too large, so that the regulation is performed once in 10 minutes in the case, and the delta t is better01、Δt12Should be less than or equal to the regulation interval, let Δ t be the most accurate combination of time intervals01And Δ t12Respectively for 2, 4, 6, 8 and 10 minutesAnd 25 working conditions in total. However, if the system parameters are stable, the system operates stably after one time of regulation, and as shown in the embodiment, the control parameters of the system are selected time by time (every 60 minutes).
Step 6, until the detection by the test sample is finished, corresponding to all different delta t01、Δt12The prediction errors of the combined prediction models are compared with the prediction precision of the prediction models of different time scales;
And (2) performing importance sequencing on the influence of outdoor meteorological parameters (including outdoor dry bulb temperature, outdoor wet bulb temperature, air relative humidity, total solar radiation illumination and wind speed) and indoor environment parameters (including indoor air temperature and indoor air relative humidity) on the load by adopting a correlation analysis method, wherein the positive numerical value represents positive correlation with the load, the negative numerical value represents negative correlation with the load, sequencing is performed according to the absolute value, and the parameter with the absolute value larger than 0.2 is selected as a main influence parameter. For the selected embodiment of the office building, the outdoor dry bulb temperature, the indoor air temperature and the outdoor air relative humidity are finally determined main influence parameters. Δ t01And Δ t12Respectively taking 10 minutes, 20 minutes, 30 minutes, 40 minutes, 50 minutes and 60 minutes, and combining for 36 minutes. When Δ t is reached01=30,Δt12At 60, the ANN prediction error is the smallest, so the ANN at this time interval combination is selected as the final load prediction model of the embodiment of the present invention.
The invention particularly selects an Artificial Neural Network (ANN) method as a method for establishing the load prediction model, and other equivalent methods can be adopted if other machine learning algorithms with higher prediction precision and simple and convenient use exist.
Δt01、Δt12The initial value of (a) is selected mainly according to the time lag of the system and the expected control precision of the air conditioning system, and is not limited to the selection method in the embodiment. If the time lag is several hours, an interval of the order of hours may be chosen. Final Δ t01,Δt12Is determined based on the prediction error.
The invention fully considers the delay of the system, respectively discusses the system control method with the delay time less than 1h or more than 1h, and effectively improves the adaptability of the air conditioning system load and the indoor thermal environment. The method can be applied to an automatic control system for optimizing the air conditioner so as to meet the requirement of the room thermal environment and reduce the energy consumption of the air conditioner.
Claims (1)
1. An air conditioner side load prediction method based on system delay is characterized by comprising the following steps:
step 1, collecting historical data of outdoor meteorological parameters, indoor thermal environment parameters and air conditioner side output loads of a target room;
step 2, setting the time scales of the past and the future, specifically: let t be the time at a certain time in the historical data collected in step 11,t1A certain time before is t0,t1A certain time later is t2Wherein t is2The indoor parameter at the moment is the control target, t0And t1Time interval Δ t of01And t1And t2Time interval Δ t of12;
Step 3, mixing t0、t1、t2Indoor and outdoor parameters and Δ t at corresponding times01、Δt12Taking the average air conditioner output load of the corresponding time interval as a sample; for t0、t1、t2Giving different values to obtain a plurality of training and testing samples for generating a prediction model;
step 4, establishing a prediction model of the air conditioner output load by using the sample obtained in the step 3: the input variable of the prediction model is t0、t1、t2Corresponding indoor and outdoor parameters and Δ t01Average air conditioner output load of corresponding time interval, output is delta t12Dividing all samples into a training sample set and a testing sample set according to the proportion of the average air conditioner output load of the corresponding time interval, training a prediction model by using the training sample set, and detecting by using the testing sample set to obtainObtaining the prediction error of the prediction model, and storing the trained prediction model and the corresponding delta t01、Δt12;
Step 5, changing delta t01、Δt12Regulation, i.e. for all deltat01、Δt12The combination adopts a traversal method to establish a model one by one, and whether the past time interval delta t is completed or not01And a future time interval Δ t12If not, returning to the step 2; if yes, go to step 6;
step 6, until the detection by the test sample is finished, corresponding to all different delta t01、Δt12The prediction errors of the combined prediction models are compared with the prediction accuracy of the prediction models under different time scales;
step 7, selecting the prediction model with the minimum error and the corresponding delta t01、Δt12And combining the two to form a final air conditioner output load prediction model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011139927.5A CN112268350B (en) | 2020-10-22 | 2020-10-22 | Air conditioner side load prediction method based on system delay |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011139927.5A CN112268350B (en) | 2020-10-22 | 2020-10-22 | Air conditioner side load prediction method based on system delay |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112268350A CN112268350A (en) | 2021-01-26 |
CN112268350B true CN112268350B (en) | 2022-03-15 |
Family
ID=74341747
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011139927.5A Active CN112268350B (en) | 2020-10-22 | 2020-10-22 | Air conditioner side load prediction method based on system delay |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112268350B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112994031A (en) * | 2021-03-19 | 2021-06-18 | 国网安徽省电力有限公司电力科学研究院 | SVM static reactive load modeling method based on air conditioner load proportion |
CN116045461B (en) * | 2023-03-07 | 2023-10-27 | 广东热矩智能科技有限公司 | Energy-saving control method and device for air-cooled air conditioner based on water supply and return temperature adjustment |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0886490A (en) * | 1994-09-14 | 1996-04-02 | Toshiba Corp | Predicting equipment of thermal load |
CN101877077A (en) * | 2009-11-25 | 2010-11-03 | 天津工业大学 | Time series predicting model |
CN108197404B (en) * | 2018-01-22 | 2021-03-16 | 河北工业大学 | Building load prediction method based on time genetic characteristics |
CN108709287B (en) * | 2018-04-03 | 2020-09-01 | 广东迪奥技术有限公司 | Air conditioning system cold load prediction method and water chilling unit group control strategy |
JP7343289B2 (en) * | 2019-03-29 | 2023-09-12 | 三機工業株式会社 | Air conditioning load prediction method and air conditioning system |
CN111735177B (en) * | 2020-07-15 | 2024-06-18 | 西安木牛能源技术服务有限公司 | Central air conditioning system cold load prediction method based on SVR algorithm |
-
2020
- 2020-10-22 CN CN202011139927.5A patent/CN112268350B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN112268350A (en) | 2021-01-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112128841B (en) | Whole-network balance adjusting method based on load prediction and room temperature feedback correction | |
Pfeiffer et al. | Control of temperature and energy consumption in buildings-a review. | |
CN109253494B (en) | Control method of electric heat storage device based on heat load prediction | |
CN102679505B (en) | Room temperature control method | |
CN106842914B (en) | Temperature control energy-saving processing method, device and system | |
CN112268350B (en) | Air conditioner side load prediction method based on system delay | |
CN110726218B (en) | Air conditioner, control method and device thereof, storage medium and processor | |
Javed et al. | Comparison of the robustness of RNN, MPC and ANN controller for residential heating system | |
CN112926795A (en) | SBO (statistical analysis) -based CNN (continuous casting) optimization-based high-rise residential building group heat load prediction method and system | |
CN112984616A (en) | Method for identifying regulation and control rule of heating power station of heating system | |
CN113587172A (en) | Water supply temperature delay time prediction method and device and electronic equipment | |
CN114046593A (en) | Dynamic predictive machine learning type air conditioner energy-saving control method and system | |
CN115907191A (en) | Adaptive building photovoltaic skin model prediction control method | |
Huang et al. | Model predictive control for energy-efficient buildings: An airport terminal building study | |
KR20200092478A (en) | Method for controlling economizer air conditioning system | |
CN114200839B (en) | Intelligent office building energy consumption control model for dynamic monitoring of coupling environment behaviors | |
CN113028610B (en) | Method and device for global optimization and energy-saving control of dynamic load of central air conditioner | |
Sha et al. | A control strategy of heating system based on adaptive model predictive control | |
KR20180138371A (en) | Method for evaluating data based models and conducting predictive control of capsule type ice thermal storage system using the same | |
EP3771957A1 (en) | Method and system for controlling of heating, ventilation and air conditioning | |
Javed et al. | Modelling and optimization of residential heating system using random neural networks | |
CN115935604A (en) | Energy-saving control method and terminal for heating and ventilation system of converter station | |
CN113625557A (en) | HVAC system model prediction control method of online optimization model | |
US20210033299A1 (en) | Method and system for controlling heating, ventilation and air conditioning | |
CN115660325B (en) | Power grid peak regulation capacity quantization method and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CP02 | Change in the address of a patent holder |
Address after: 300452 Binhai Industrial Research Institute Campus of Tianjin University, No. 48 Jialingjiang Road, Binhai New Area, Tianjin Patentee after: Tianjin University Address before: 300072 Tianjin City, Nankai District Wei Jin Road No. 92 Patentee before: Tianjin University |
|
CP02 | Change in the address of a patent holder |