CN117193034B - Building intelligent control method and system - Google Patents
Building intelligent control method and system Download PDFInfo
- Publication number
- CN117193034B CN117193034B CN202311465113.4A CN202311465113A CN117193034B CN 117193034 B CN117193034 B CN 117193034B CN 202311465113 A CN202311465113 A CN 202311465113A CN 117193034 B CN117193034 B CN 117193034B
- Authority
- CN
- China
- Prior art keywords
- state
- distribution
- states
- building
- probability
- 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
- 238000000034 method Methods 0.000 title claims abstract description 36
- 230000007704 transition Effects 0.000 claims abstract description 112
- 238000010438 heat treatment Methods 0.000 claims abstract description 44
- 238000001816 cooling Methods 0.000 claims abstract description 36
- 238000011217 control strategy Methods 0.000 claims abstract description 35
- 230000008859 change Effects 0.000 claims abstract description 8
- 238000009826 distribution Methods 0.000 claims description 117
- 239000011159 matrix material Substances 0.000 claims description 59
- 230000006399 behavior Effects 0.000 claims description 23
- 238000010606 normalization Methods 0.000 claims description 10
- 239000002131 composite material Substances 0.000 claims description 7
- 230000000295 complement effect Effects 0.000 claims description 5
- 230000015572 biosynthetic process Effects 0.000 claims description 3
- 238000003786 synthesis reaction Methods 0.000 claims description 3
- 230000002123 temporal effect Effects 0.000 claims description 3
- 238000005057 refrigeration Methods 0.000 claims description 2
- 238000004422 calculation algorithm Methods 0.000 abstract description 6
- 238000007405 data analysis Methods 0.000 abstract description 4
- 230000007547 defect Effects 0.000 abstract description 2
- 238000004378 air conditioning Methods 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 238000012986 modification Methods 0.000 description 7
- 230000004048 modification Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 238000005265 energy consumption Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000009472 formulation Methods 0.000 description 4
- 239000000203 mixture Substances 0.000 description 4
- 238000007796 conventional method Methods 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 210000004556 brain Anatomy 0.000 description 2
- 239000008358 core component Substances 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000001932 seasonal effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010924 continuous production Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000009423 ventilation Methods 0.000 description 1
- 239000011800 void material Substances 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Abstract
The invention discloses a building intelligent control method and system; step1: defining a state set: the invention first defines a set of states of a building system, each state representing a state or a combination of states of the building system at a time step. This is the basis of the method of the invention, since the markov chain will transition according to these states. The states or state combinations S include a cooling mode (C), a heating mode (H) and a shut-down mode (O), the state sets being expressed as:step2: markov chain: in this step, the present invention creates a Markov chain to describe transitions between states; 1. flexibility and adaptivity: the intelligent control method of the invention overcomes the defects of fixed rules and logic in the prior art. Through real-time data analysis and intelligent algorithm, the system can automatically adjust the control strategy according to the current dynamic change conditions of environment, season, personnel flow and the like.
Description
Technical Field
The invention relates to the technical field of building automation, in particular to an intelligent control method and system for a building.
Background
Building automation refers to the use of advanced technologies and systems to enable automated control and management of building internal devices, systems and processes. The aim is to improve the energy efficiency, safety, comfort and operation efficiency of the building. Conventional building automation typically includes monitoring, controlling, and optimizing various building systems, such as heating, ventilation, air conditioning, lighting, security systems, and the like.
Conventional building automation intelligent control methods generally employ control systems based on hardware sensors and program logic. These systems control the operation of the individual systems by collecting sensor data, such as temperature, humidity, illumination, etc., and then according to predetermined rules and logic. For example, the switching of a lighting, air conditioning or heating system is automatically controlled according to a schedule or a specific event.
However, the following technical problems are needed to be solved in the conventional technology:
(1) Fixed rules and logic: conventional approaches typically control based on predefined rules and logic, lacking flexibility and adaptation. They cannot accommodate complex changes and unknowns.
(2) The method can not adapt to dynamic changes: the traditional method is difficult to cope with dynamic changes of building environment and demands, such as seasonal changes, changes of personnel flow, and the like.
(3) No comprehensive analysis: conventional methods typically control with each system as an independent unit, lacking comprehensive analysis between the different systems.
These drawbacks are mainly due to the rigid and static nature of the traditional methods. Traditional methods rely on fixed rules and hardware sensors, lacking real-time analysis of data and intelligent decision making capability. In addition, the conventional method often has no complete data integration and system interconnection, so that the systems cannot work cooperatively, and the improvement of the overall performance is limited.
Therefore, an intelligent building control method and system are provided.
Disclosure of Invention
In view of the above, the embodiments of the present invention provide a building intelligent control method and system, so as to solve or alleviate the technical problems existing in the prior art, that is, control design of rigidity and static properties, and provide at least one beneficial choice for the control design;
the technical scheme of the embodiment of the invention is realized as follows:
aspect 1
An intelligent control method for a building comprises the following steps:
step1: defining a state set: the invention first defines a set of states of a building system, each state representing a state or a combination of states of the building system at a time step. This is the basis of the method of the invention, since the markov chain will transition according to these states. The states or state combinations S include a cooling mode (C), a heating mode (H) and a shut-down mode (O), the state sets being expressed as: 。
Step2: markov chain: in this step, the present invention creates a Markov chain to describe transitions between states.
Specifically:
step2.1: the invention defines the transition probability under the given current state through the transition probability matrix. This will determine the probability that the system will transition to the next state given the current state. The transition probability matrix P:
the transition probability is: in the current time step, the state is the cooling mode (C), and the state of the next time step is the cooling mode (C) or the heating mode (H), but it is impossible to be the off mode (O).
Step2.2: the present invention defines markov properties and initial state distributions. Markov properties indicate that future states depend only on current states, independent of past states. The initial state distribution represents a probability distribution of each state at the beginning of the initial point in time.
In STEP-2.2, the Markov property and the initial state distribution are in a cooling mode with 50% probability, a heating mode with 30% probability and a closing mode with 20% probability;
the state distribution of the building system is:
the initial state distribution is at an initial node of the Markov chain.
Step2.3: the invention ensures the satisfaction of transition probabilities, i.e. they are non-negative and normalized. This is the basic property of ensuring probability. In the transition probability matrix P, P (i, j) represents the probability of transition from state i to state j:
the non-negativity: for all i and j, P (i, j) is more than or equal to 0;
the normalization: for each state i, Σp (i, j) =1 is satisfied, where Σ is the sum of all possible states j;
the transition probability is between 0 and 1:
for all i and j, 0.ltoreq.P (i, j). Ltoreq.1 is satisfied.
Step2.4: the invention defines temporal homogeneity, meaning that the transition probability remains unchanged over time. This is a key feature of the Markov chain, so that the invention can model the evolution of states as a continuous process. The time-uniformity: for any time step t and t+1, the transition probability matrix remains unchanged:
indicating that at any one time step, the probability of transition from one state to another is fixed and does not change over time.
Step2.5: finally, the present invention uses a Markov chain to predict building system state behavior for the next time step. In STEP-2.5, based on the current state distribution and transition probability, predicting the state distribution of the next time STEP by matrix multiplication:
The state vector P (t) represents the distribution of the state or combination of states S at the current time step:
wherein P (Si, t) is the probability of the state Si at time step t;
the matrix multiplication:
matrix multiplication multiplies the current state distribution vector by a transition probability matrix to obtain a state distribution vector of the next time step, and each element P (Si, t+1) represents the probability of the state Si at time step t+1.
Step3: D-S evidence theory at this step, the present invention uses the D-S evidence theory to verify the predicted state behavior and output the final control strategy. D-S evidence theory is a method for processing uncertainty and merging evidence, and the invention compares predicted state behavior with actual observed state behavior and then makes a final control decision according to the reliability of the evidence; in STEP-3, the predicted state behavior is the state vector P (t+1);
two evidence items are provided, which respectively represent a predicted state distribution vector P (t+1) and an actually observed state distribution vector O, wherein O is expressed as:
combining the two evidences using D-S evidence theory, resulting in a final confidence distribution vector C, where C (i) represents the degree of belief of the state Si, D-S synthesis formula:
Where Bel (A) is the base confidence level, pl (A) is the average confidence level, and U (A) is the uncertainty measure.
For an evidence item A, the basic trust level is calculated:
where BelT (a) represents the confidence in setting a under different evidence sources, this summation covering all evidence sources;
calculate the average confidence level Pl (a):
where ¬ A represents the complement of A and Bel (¬ A) represents the basic confidence that ¬ A is not believed;
uncertainty metric U (A) represents the strength of confidence in evidence:
calculate the resultant confidence level (CA):
where U (¬ A) represents the uncertainty measure of ¬ A and the resultant confidence level represents the final confidence level for setting A.
Aspect 2
An intelligent building control system, comprising:
(1) Processor (Processor): this is the core component of the system, responsible for executing program instructions, performing data processing and decision making. It is the brain of the system that analyzes the data collected from the various sensors according to predefined logic and algorithms to control the operation of the various systems within the building.
(2) Memory (Memory): the memory is where program instructions, data, and intermediate results are stored. It stores various information required for the operation of the drive system, including building state sets, transition probability matrices, weight functions, etc.
(3) Program instructions (ProgramInstructions): the memory stores a series of program instructions that define how the system makes decisions and controls based on the collected data. These instructions may range from data processing to state prediction to final control strategy formulation.
The system work flow:
p1, data collection: a sensor group of conventional building automation is incorporated into the driving range of the control system to collect various data in the building, such as temperature, humidity, personnel flow, etc. These data are passed to the processor.
P2, data analysis: the processor analyzes and processes the collected data according to predefined program instructions in the memory. For example, the state distribution of the next time step is predicted from the transition probability matrix.
P3, intelligent decision: and the processor establishes a proper control strategy by utilizing an intelligent algorithm and decision logic according to the analysis result. This may include adjusting the operating state of the lighting, air conditioning, heating, etc. system.
P4, control execution: the processor converts the formulated control strategy into a control instruction, and realizes real-time control of the building system by communicating with the interfaces of the systems.
Compared with the prior art, the invention has the beneficial effects that:
1. flexibility and adaptivity: the intelligent control method of the invention overcomes the defects of fixed rules and logic in the prior art. Through real-time data analysis and intelligent algorithm, the system can automatically adjust the control strategy according to the current dynamic change conditions of environment, season, personnel flow and the like. This flexibility and adaptation enables the system to respond more accurately to changes, thereby improving building efficiency and comfort.
2. Adapting to dynamic changes: the traditional technology is difficult to adapt to the dynamic change of the internal and external environments of the building. The intelligent control method fully utilizes real-time data and can carry out real-time adjustment according to the changes of temperature, humidity, personnel flow and the like. The system can react quickly, whether seasonally or temporarily, to provide a more optimal control strategy.
3. Comprehensive analysis: conventional techniques typically control individual building systems in a decentralized manner. The intelligent control method can realize comprehensive analysis of a plurality of systems by integrating various data and information. Such comprehensive analysis helps to formulate a more comprehensive control strategy and improves overall operation efficiency.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required in the embodiments or the technical descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of the method logic of the present invention;
FIG. 2 is a control program diagram of embodiment 7 of the present invention;
FIG. 3 is a control program diagram of embodiment 7 of the present invention;
FIG. 4 is a control program diagram of embodiment 7 of the present invention;
FIG. 5 is a schematic diagram of the control system of the present invention.
Detailed Description
In order that the above objects, features and advantages of the invention will be readily understood, a more particular description of the invention will be rendered by reference to the appended drawings. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. The present invention may be embodied in many other forms than described herein and similarly modified by those skilled in the art without departing from the spirit of the invention, whereby the invention is not limited to the specific embodiments disclosed below.
Example 1
As shown in fig. 1, a building intelligent control method comprises:
STEP-1: defining a set of states for a building system: each state represents a state or combination of states of the building system at a time step; specifically, in building intelligent control, each state represents a particular state or combination of states of a building system at a certain time step. These states may encompass various system parameters, with the purpose of defining a set of states in order to divide the complex state space of a building into manageable states.
STEP-2: markov chain:
a markov chain is a mathematical model describing transition probabilities between states. In building intelligent control, the use of markov chains is to predict future state behavior:
STEP-2.1: at this step, the present embodiment will analyze the building state set based on historical data and domain knowledge and derive the transition probabilities for a given current state. This may be achieved by statistical methods or data modeling techniques. The transition probability defines the likelihood that the system will transition to other states in the current state, providing a basis for future predictions.
STEP-2.2: markov properties indicate that future states depend only on current states, independent of previous states. This allows the present embodiment to predict future state behavior by giving the current state and transition probabilities. The initial state distribution represents the probability distribution of each state at the beginning of the initial point in time, providing a starting point for the prediction process.
STEP-2.3: the purpose of this step is to ensure that the transition probabilities meet the non-negative and normalized conditions. Each transition probability must be equal to or greater than zero and the sum of the transition probabilities for each state must be equal to 1. This ensures that transitions between states are reasonable and complete and that no probability inconsistencies occur.
STEP-2.4: defining time uniformity: the time-alignment setting indicates that the transition probability remains unchanged over time. This means that at any point in time, the transition probabilities between states remain consistent. This setting allows the present embodiment to use the same transition probability matrix at each time step, simplifying the establishment and calculation of the predictive model.
STEP-2.5: outputting predicted state behavior: based on the Markov chain model established in the previous step, the embodiment can predict the building system state behavior of the next time step. The method is realized through matrix multiplication and transition probability, and the state distribution vector of the next time step is obtained by multiplying the state distribution vector of the current time step by the transition probability matrix, so that the state prediction is realized.
STEP-3: D-S evidence theory: in this step, the embodiment verifies the predicted state behavior and the actual building system state as two evidences, and then outputs the final control strategy according to the reliability of the evidences. D-S evidence theory allows the present embodiment to combine different evidence to arrive at a final control decision. This step involves the calculation of weight functions, basic confidence levels, average confidence levels, and uncertainty metrics, ensuring that the final control strategy is more accurate and reliable.
The above examples merely illustrate embodiments of the invention that are specific and detailed for relevant practical applications and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention.
Example 2
In STEP-1, the state or combination of states S includes a cooling mode (C), a heating mode (H), and a shutdown mode (O), with the set of states expressed as:
specifically, in building intelligent control, the definition of a state set is the basis for building the whole intelligent control system. Each state represents a mode of operation of the building system under specific conditions, such as different temperature demands, seasonal variations, etc. The states can be switched in actual operation, and the control system needs to take corresponding measures according to different states so as to realize optimal energy efficiency and comfort.
For example, in the cooling mode, the system may adjust the air conditioning system to reduce the indoor temperature. In the heating mode, the system activates the heating system to raise the indoor temperature. In the off mode, the system may be shut down to save energy. The switching of these states may be determined by factors such as schedule, personnel flow, outdoor temperature, etc.
It will be appreciated that in this embodiment, the definition of the state set provides a basis for subsequent steps, such as the creation of a Markov chain model and the calculation of transition probabilities. In building intelligent control, the embodiment hopes that the system can predict future state behaviors according to current state and historical data and formulate an optimal control strategy according to a prediction result. Thus, by intelligently switching and controlling different states, the building can realize more efficient, comfortable and intelligent operation.
Further, in practice, defining the cooling mode, heating mode, and shutdown mode involves the operational status and behavior of the building system, and these definitions need to be determined according to specific building characteristics, requirements, and technologies:
(1) Cooling mode (CoolingMode):
the cooling mode is typically used in seasons where the temperature is high (e.g., summer), in order to reduce the indoor temperature. In defining the cooling mode, the present embodiment considers the following factors:
temperature threshold: a temperature threshold defining when to initiate the cooling mode. For example, when the indoor temperature exceeds a certain threshold (e.g., 25 degrees celsius), the system may switch to the cooling mode.
System equipment: the cooling mode is considered to relate to the operation of the air conditioning system. Defining when to start and stop the air conditioner and how to adjust the temperature setting of the air conditioner.
Energy efficiency considerations: in the cooling mode, the system may optimize the operation of the air conditioner to increase energy efficiency. Defines how to adjust the power and the operation mode of the air conditioner according to the indoor and outdoor temperatures and the flow of people.
(2) Heating mode (heating mode):
the heating mode is generally used in seasons where the temperature is low (e.g., winter), in order to raise the indoor temperature. In defining the heating mode, the following factors may be considered:
temperature threshold: a temperature threshold defining when to initiate the heating mode. For example, when the indoor temperature is below a certain threshold (e.g., 20 degrees celsius), the system may switch to the heating mode.
Heating system: defining when to start and stop the heating system and how to adjust the temperature settings of the heating.
Energy efficiency considerations: in the heating mode, the system can optimize the operation of the heating system to improve energy efficiency. Defines how to adjust the power and the working mode of the heating according to the indoor and outdoor temperature and the personnel flow.
(3) Off mode (OffMode):
the off mode refers to the system being in a dormant or inactive state to save energy. In defining the off mode, the following factors may be considered:
schedule: defining when to enter the off mode, for example at night or when no person is present.
System equipment: in the off mode, the system should consider stopping operation or entering a low power state. Defining how to safely shut down the various systems to avoid energy waste.
Person perception: consider how to determine whether a system needs to be started by a sensor or a person identifying the system. For example, if no person enters the building for a period of time, the switch to the off mode may be automatic.
The above examples merely illustrate embodiments of the invention that are specific and detailed for relevant practical applications and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention.
Example 3
In STEP-2.1, the transition probability is: at the current time step, the state is the cooling mode (C), and the state of the next time step may be the cooling mode (C) or the heating mode (H), but may not be the off mode (O).
The present embodiment defines the state transition probabilities for a given current state by means of a transition probability matrix. The transition probability matrix P is a square matrix, where P (i, j) represents the probability of transition from state i to state j. Transition probability matrix P:
This matrix illustrates the state possibilities of the next time step in the different states of the current time step.
Illustratively, if the current state is cooling mode (C), then at the next time step, there is a 60% probability that it is still in cooling mode and a 40% probability that it is transitioning to heating mode. But it is not possible to directly transition from the cooling mode to the off mode. Similarly, transition probabilities for other states are defined accordingly.
In STEP-2.2, the Markov property and initial state distribution are such that 50% of the probability is in the cooling mode, 30% of the probability is in the heating mode, and 20% of the probability is in the off mode; in particular, markov properties indicate that future states depend only on current states, independent of previous states. The initial state distribution represents a probability distribution of each state at the start of the initial point in time.
In particular, the system satisfies the markov property, i.e. the future state depends only on the current state, independent of the past state.
The state distribution of the building system is:
the initial state distribution is at the initial node of the Markov chain. This means that at the initial node of the markov chain, there is a 50% probability of being in the cooling mode, a 30% probability of being in the heating mode, and a 20% probability of being in the off mode.
Specifically, the initial state distribution reflects the operational state probability distribution of the building at the initial point in time. This distribution, based on historical data and a priori knowledge, may help predict the state evolution of the building at future time steps. For example, in the example provided by the present embodiment, 50% probability is in the cooling mode, 30% probability is in the heating mode, and 20% probability is in the off mode, which can be used to predict the state of the system at the next time step.
It will be appreciated that other probabilistic prediction modes may also be defined by itself based on historical data and prior knowledge.
Specifically, the initial state distribution is also significant for the formulation of the initial policy. The initial state of the system affects the initial operational behavior, which facilitates proper adjustments and decisions at start-up to meet building requirements and energy efficiency requirements.
Further, after the system enters the running state, the actually observed state distribution is compared with the initial state distribution, so that the running condition of the system can be verified. If the actual state distribution differs significantly from the initial distribution, this may mean that the system is problematic and requires further adjustment or maintenance. The initial state distribution may also be used as feedback information to optimize the control strategy. The system can monitor the change of state distribution in real time and adjust the control strategy according to actual conditions so as to adapt to the changing requirements of building operation.
It will be appreciated that in this embodiment:
the initial state distribution has decision function in the whole building intelligent control method:
(1) And (3) starting a decision: the initial state distribution affects the initial operating state and policies of the system at start-up. Different initial state distributions may require different startup strategies to reach the desired operating mode as soon as possible.
(2) Initial control: according to the initial state distribution, the system can determine whether certain systems need to be started in advance or not so as to ensure that the building can meet the requirements in the initial operation stage.
(3) State prediction: the initial state distribution provides a starting point for the Markov chain model. The initial state distribution is one of the predicted initial conditions when predicting the state of the next time step.
In this embodiment, STEP-2.3 is a key part of the building intelligent control, which describes the transition probabilities between the system states. To ensure the rationality and usability of this matrix, the present embodiment needs to meet the following conditions:
(1) Non-negativity: the transition probabilities P (i, j) need to satisfy the non-negativity, i.e., P (i, j). Gtoreq.0, for all states i and j. This is because the transition probability is a probability value and must be zero or more.
(2) Normalization: for each state i, the sum of each row of the transition probability matrix needs to be equal to 1, i.e. Σp (i, j) =1, where Σ is the sum of all possible states j. This ensures that at each time step the system must transition to a state without loss of state or repeated transitions.
(3) Transition probabilities between 0 and 1: the transition probability P (i, j) needs to satisfy 0.ltoreq.P (i, j). Ltoreq.1 for all states i and j. This is because the transition probability represents the probability of a transition from state i to state j, which must be between 0 and 1.
These conditions ensure the rationality and interpretability of the transition probability matrix. The requirement that the non-negativity and transition probabilities are between 0 and 1 ensures the validity of the probabilities, while the normalization requirement ensures the completeness and mutual exclusivity between the states. In this way, the present embodiment can correctly perform state prediction and transition in the markov chain model.
Illustratively, each element satisfies the requirements of non-negativity, normalization, and transition probabilities between 0 and 1. Such a matrix may be used to predict the state of the next time step given the current state.
By way of example, this embodiment provides a building system for an office building that includes three states, a cooling mode (C), a heating mode (H), and an off mode (O). This embodiment will show how to construct a transition probability matrix, define an initial state distribution, and verify that the conditions of non-negativity, normalization, and transition probability between 0 and 1 are satisfied:
Transition probability matrix P:
initial state distribution:
STEP STEP-2.1: defining a transition probability matrix:
according to an example scenario, the present embodiment defines the transition probability matrix described above. This matrix describes the state possibilities for the next time step in different states. For example, the probability of transition from the cooling mode (C) to the heating mode (H) is 0.3.
STEP-2.2 defining markov properties and initial state distribution:
according to an example scenario, the present embodiment defines an initial state distribution. At the beginning of the initial time point, there is a 40% probability of being in the cooling mode, a 30% probability of being in the heating mode, and a 30% probability of being in the off mode. This initial state distribution will act as the initial node of the markov chain.
STEP STEP-2.3, defining satisfaction of transition probability:
according to the transition probability matrix and the initial state distribution, the embodiment can verify the satisfaction condition of the transition probability:
non-negativity: in the example of the present embodiment, all transition probabilities are non-negative, satisfying the non-negative condition.
Normalization: for each row of the transition probability matrix, the sum of the individual elements should be equal to 1. In an example, for each row, e.g., the first row (state C), 0.6+0.3+0.1=1, the normalization condition is satisfied.
Transition probabilities between 0 and 1: all transition probabilities are between 0 and 1, and the condition of the value range of the transition probabilities is met.
By way of the above example, the present embodiment demonstrates how to construct a transition probability matrix according to a specific scenario, define an initial state distribution, and verify that conditions of non-negativity, normalization, and transition probability between 0 and 1 are satisfied. The steps provide a foundation for the establishment of the Markov chain model, so that the state behaviors of the building system are predicted at different time steps, and support is provided for intelligent control decisions.
The above examples merely illustrate embodiments of the invention that are specific and detailed for relevant practical applications and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention.
Example 4
STEP-2.4, time-alignment: for any time step t and t+1, the transition probability matrix remains unchanged:
indicating that at any one time step, the probability of transition from one state to another is fixed and does not change over time. Temporal uniformity is to reflect the stability and consistency of the system. In building intelligent control, the transition probability matrix is set to be unchanged with time, so that the embodiment can predict state distribution by using the same transition probability at different time steps, and the complexity of a model is simplified.
Further, in building intelligent control, time uniformity indicates that the state transition probability of the system remains unchanged at different time steps. This means that the probability of transition from one state to another is fixed, not changing over time, either at time step t or at time step t+1. This property is true in many practical situations because the operating laws of building systems are generally relatively stable and do not fluctuate drastically over time.
In STEP-2.5, the state distribution of the next time STEP is predicted by matrix multiplication based on the current state distribution and transition probability:
the state vector P (t) represents the distribution of states or state combinations S at the current time step:
where P (Si, t) is the probability of the state Si at time step t.
Matrix multiplication:
matrix multiplication multiplies the current state distribution vector by a transition probability matrix to obtain a state distribution vector of the next time step, and each element P (Si, t+1) represents the probability of the state Si at time step t+1. In order to predict the state distribution of the next time step, the present embodiment multiplies the state distribution vector P (t) of the current time step by the transition probability matrix P to obtain the state distribution vector P (t+1) of the next time step. This is an operation of matrix multiplication, where P (Si, t+1) represents the probability of state Si at time step t+1.
Further, the present embodiment can combine the state distribution of the current time step with the transition probability by matrix multiplication, so as to predict the state distribution of the system in the next time step. The prediction can be continuously performed in the running process of the system, so that the embodiment is helped to know the change trend of the system state and make corresponding control decisions.
Further, in STEP-2.5, the present embodiment predicts the state distribution of the next time STEP by matrix multiplication using the current state distribution and transition probability matrix. The state vector P (t) represents the distribution of states or state combinations S at the current time step, i.e. the probability of each state. The matrix P represents a transition probability matrix, where P (si|sj) represents the probability of transition from state Sj to state Si.
Illustratively, this embodiment extends over the scenario of embodiment 3:
STEP STEP-2.4: time-alignment:
under the setting of time uniformity, the transition probability matrix remains unchanged, i.e., P (t) =p (t+1) =p.
STEP STEP-2.5: predicting the state distribution of the next time step:
the present embodiment will use the initial state distribution and the transition probability matrix for matrix multiplication to predict the state distribution for the next time step.
First, the present embodiment constructs a state vector P (t):
then, the present embodiment builds a transition probability matrix P, where the present embodiment uses the previous example transition probability matrix:
next, the present embodiment performs matrix multiplication to obtain a state distribution vector P (t+1) of the next time step:
/>
this means that at the next time step the state profile of the system will become 32% in cooling mode (C), 36% in heating mode (H), 32% in shut-down mode (O). Through the steps of time uniformity and state prediction, the embodiment predicts the state distribution of the system in the next time step by using the transition probability matrix and the initial state distribution. This prediction can be used in the formulation of intelligent control strategies to help building systems make reasonable decisions based on past state evolution to accommodate changing environments and demands.
The above examples merely illustrate embodiments of the invention that are specific and detailed for relevant practical applications and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention.
Example 5
In STEP-3, the D-S evidence theory is used to combine the predicted state behavior and the actual observed state distribution vector to obtain a final confidence distribution vector. The purpose of this step is to rationally fuse the information from the different sources in order to make more reliable control decisions.
D-S evidence theory, also known as Dempster-Shafer evidence theory, is a method for merging incomplete or uncertain information. In building intelligent control, the embodiment needs to combine the predicted state distribution vector and the actually observed state distribution vector to obtain a more accurate state confidence distribution vector.
Predicting state behavior as a state vector P (t+1);
two evidence items are provided, which respectively represent a predicted state distribution vector P (t+1) and an actually observed state distribution vector O, wherein O is expressed as:
combining the two evidences using D-S evidence theory, resulting in a final confidence distribution vector C, where C (i) represents the degree of belief of the state Si, D-S synthesis formula:
where Bel (A) is the base confidence level, pl (A) is the average confidence level, and U (A) is the uncertainty measure.
The basic trust Bel (A) represents the degree of trust in setting A. Its calculation involves summing the trust level for setting a under all evidence sources. This calculation takes into account the information provided by all evidence sources to construct an initial confidence profile.
For an evidence item A, the basic trust level is calculated:
where BelT (a) represents the confidence in setting a under different evidence sources, this summation covers all evidence sources.
Calculate the average confidence level Pl (a):
where ¬ A represents the complement of A and Bel (¬ A) represents the basic confidence that ¬ A is not believed; the average degree of trust Pl (a) represents the degree of distrust on the opposite (complement ¬ a) of setting a. By subtracting the base confidence Bel (¬ A) for ¬ A from 1, the distrust for ¬ A can be obtained. Such calculations take into account the attitudes to the opposite settings so that the overall trust profile is balanced.
Uncertainty metric U (A) represents the strength of confidence in evidence:
uncertainty metric U (A) represents the trust strength for evidence. By subtracting the average confidence level Pl (a) from 1, this embodiment can obtain the uncertainty level for the evidence. This measure reflects the level of trust that the present embodiment has in the evidence, with higher indicating that the present embodiment is less certain of its reliability.
Calculate the resultant confidence level (CA):
where U (¬ A) represents the uncertainty measure of ¬ A and the resultant confidence level represents the final confidence level for setting A. The resultant confidence level C (A) is based on the product of the basic confidence level Bel (A) and the uncertainty measure U (¬ A). This calculation takes into account the effect of uncertainty, if the uncertainty for the opposite setting ¬ a is higher, the confidence in a will decrease accordingly.
Further, in this embodiment, the predicted state behavior and the actually observed state distribution are combined, so as to obtain a final confidence distribution vector. This confidence distribution vector reflects the degree of trust for each state, helping this embodiment make more reliable decisions in the intelligent control.
Illustratively, let us say that the present embodiment has predicted the state distribution vector P (t+1) for the next time STEP by STEP-2.5, while the actual observed state distribution vector O is as follows:
actual observed state distribution vector O:
the present embodiment will now use D-S evidence theory to combine the predicted state behavior and the actual observed state distribution vector to obtain a final confidence distribution vector.
STEP STEP-3, information fusion is carried out by using a D-S evidence theory:
first, the present embodiment calculates a basic confidence Bel (a) and an average confidence Pl (a). Set ¬ a to a current state is not state Si, then its complement ¬ a indicates that the current state is state Si.
Where BelT (Ai) represents the confidence in setting a under different evidence sources, the present embodiment will employ a relatively uniform confidence profile, setting each BelT (Ai) to 0.33 (this is just an example value, which may be adjusted in practice according to circumstances).
Then, the average confidence level Pl (a) is calculated:
where Bel (¬ Ai) represents a basic confidence that ¬ a is not believed, the present embodiment likewise employs a relatively uniform confidence profile, setting each Bel (¬ Ai) to 0.33.
Next, an uncertainty measure U (a) is calculated:
where Pl (Ai) represents the average confidence in ¬ Ai, again with a relatively uniform confidence profile, each Pl (Ai) is set to 0.33.
Finally, the synthesized confidence level C (A) is calculated:
this sample embodiment results in a final confidence distribution vector C in which each element represents a final degree of trust for the corresponding state.
The above examples merely illustrate embodiments of the invention that are specific and detailed for relevant practical applications and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention.
Example 6
In STEP-3, after the resulting trust level distribution vector C (a) is obtained, this distribution can be used to formulate the final control strategy. The final control strategy is to make decisions based on the trust degree of the synthesized trust degree distribution to different states, so that intelligent control of the building system is realized.
Outputting a final control strategy:
(1) Selection of a threshold: in the composite confidence distribution vector, a threshold may be selected to determine which states are considered trusted states and which states may not be employed. The selection of this threshold may be based on practical circumstances, requiring comprehensive consideration of the distribution of trust and the requirements of the system.
(2) And (3) formulating a control strategy: and according to the selected threshold value, the state exceeding the threshold value in the composite trust degree distribution is regarded as a trusted state, and the state not exceeding the threshold value is regarded as an untrusted state. Based on this determination, different control strategies may be formulated.
For example, if the confidence level of the cooling mode (C) is highest, it may be selected to switch to the cooling mode at the next time step; if the confidence level is high, the heating mode (H) is selected to switch to the heating mode.
(3) The final control strategy for executing the control strategy is formulated according to the composite trust level distribution, and the system executes corresponding operations according to the strategy, such as adjusting the temperature control system of the building, switching on or off the refrigeration equipment, the heating equipment and the like, so as to realize the selected state.
Further, the final control strategy: the final control strategy is an operation plan for the building system, selecting the most appropriate state or combination of states based on the composite confidence profile. This strategy can instruct the building intelligent control system what to take at the next time step to optimize the energy consumption, comfort and efficiency of the building.
Exemplary:
(1) The energy consumption is reduced: in building intelligent control, the composite trust distribution vector can guide the system to select the most suitable state or operation so as to reduce energy consumption. For example, if the trusted status is a cooling mode, the system may adjust the air conditioning system based on this information to reduce energy consumption and keep the building within a suitable temperature range.
(2) Comfort is improved: the synthesized trust level distribution vector considers the trust levels of different states, and can ensure that the comfort of the user is fully considered when making a decision. For example, if the trusted state is a heating mode, the system may choose to switch to the heating mode to provide a more comfortable indoor environment.
(3) And (3) improving system efficiency: the control strategy formulated based on the composite confidence profile may help optimize the overall efficiency of the building system. The system decides the state conversion according to the trust degree, thereby realizing more reasonable equipment operation and energy consumption management. This helps to avoid unnecessary energy waste and improves the operating efficiency of the system.
The above examples merely illustrate embodiments of the invention that are specific and detailed for the relevant practical applications, but are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention.
Example 7
Fig. 4 is a control program diagram of a seventh embodiment of the present invention.
Referring to fig. 5, a building intelligent control system includes:
(1) Processor (Processor): this is the core component of the system, responsible for executing program instructions, performing data processing and decision making. It is the brain of the system that analyzes the data collected from the various sensors according to predefined logic and algorithms to control the operation of the various systems within the building.
(2) Memory (Memory): the memory is where program instructions, data, and intermediate results are stored. It stores various information required for the operation of the drive system, including building state sets, transition probability matrices, weight functions, etc.
(3) Program instructions (ProgramInstructions): the memory stores a series of program instructions that define how the system makes decisions and controls based on the collected data. These instructions may range from data processing to state prediction to final control strategy formulation.
The system work flow:
p1, data collection: a sensor group of conventional building automation is incorporated into the driving range of the control system to collect various data in the building, such as temperature, humidity, personnel flow, etc. These data are passed to the processor.
P2, data analysis: the processor analyzes and processes the collected data according to predefined program instructions in the memory. For example, the state distribution of the next time step is predicted from the transition probability matrix.
P3, intelligent decision: and the processor establishes a proper control strategy by utilizing an intelligent algorithm and decision logic according to the analysis result. This may include adjusting the operating state of the lighting, air conditioning, heating, etc. system.
P4, control execution: the processor converts the formulated control strategy into a control instruction, and realizes real-time control of the building system by communicating with the interfaces of the systems.
Further, referring to fig. 2-3, exemplary program instructions are shown, which show only the logic in the form of c++ pseudo code, and the principle is as follows:
(1) num State, an enumerated type State is defined herein, comprising three states: cooling mode, heating mode, and off mode. This enumeration type is used to represent different states of the building system.
(2) double transitionMatrix [3] [3]: this is a 3*3 transition probability matrix representing transition probabilities between different states. For example, transitionMatrix [0] [1] represents the probability of transitioning from cooling mode to heating mode.
(3) double initialDistribution [3] this is an initial state distribution vector representing the probability distribution of each state at an initial point in time.
(4) vector (3, 0.0) this is a vector storing the trust distribution, with an initial value of 0.0.
(5) void combineTrust (vector "& TrustDist, double threshold) this is a function of the synthetic confidence distribution vector. In practical applications, the trust level of different states needs to be calculated according to the D-S evidence theory.
(6) main (): the main function part contains a process that simulates the actual operation. First, from predictions and observations, a confidence distribution vector is synthesized using the combinertrust function. And then, determining a final control strategy by comparing the trust level, and outputting a final control strategy result.
The above examples merely illustrate embodiments of the invention that are specific and detailed for the relevant practical applications, but are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention.
Claims (4)
1. The intelligent building control method is characterized by comprising the following steps of:
STEP-1: defining a set of states for a building system: each state represents a state or combination of states of the building system at a time step;
STEP-2: markov chain:
STEP-2.1: a state set is solved, and transition probability under a given current state is defined through a transition probability matrix;
STEP-2.2: defining Markov properties and initial state distribution of a Markov chain, wherein the initial state distribution is probability distribution of each state at the beginning of an initial time point;
STEP-2.3: defining satisfaction of transition probabilities, wherein the transition probabilities satisfy non-negativity and normalization, each transition probability is between 0 and 1, and the sum of the transition probabilities of each state is equal to 1;
STEP-2.4: defining time uniformity: the transition probability of the Markov chain is kept unchanged in time, and the transition probability among states is kept consistent at any time point;
STEP-2.5: outputting predicted state behavior: outputting predicted state behaviors of the building system at the next time step;
STEP-3: D-S evidence theory: receiving the predicted state behavior, verifying the predicted state behavior and a state set of the building system as two evidences, and outputting a final control strategy;
In STEP-1, the state or combination of states S includes a cooling mode C, a heating mode H, and a closing mode O, the set of states being expressed as:;
in STEP-2.1, the transition probability matrix P:the transition probability is: at the current time step, the state is a cooling mode C, and the state of the next time step is a cooling mode C or a heating mode H;
in STEP-2.4, the temporal uniformity: for arbitrary time steps t and t+1, the transition probability matrixRemain unchanged:
indicating that at no time step, the probability of transition from one state to another is fixed and does not change over time;
in STEP-2.5, the state distribution of the next time STEP is predicted by matrix multiplication based on the current state distribution and transition probability:
the state vector P (t) represents the distribution of the state or combination of states S at the current time step:
wherein P (Si, t) is the probability of the state Si at time step t;
the method comprises the steps of carrying out a first treatment on the surface of the The matrix multiplication: />Matrix multiplication multiplies the current state distribution vector by a transition probability matrix to obtain a state distribution vector of the next time step, and each element P (Si, t+1) represents the probability of the state Si at the time step t+1;
In STEP-3, the predicted state behavior is the state vector P (t+1);
two evidence items are provided, which respectively represent a predicted state distribution vector P (t+1) and an actually observed state distribution vector O, wherein O is expressed as:
combining the two evidences using D-S evidence theory, results in a final confidence distribution vector C, where C (Si) represents the degree of belief of the state Si, D-S synthesis formula:
where Bel (A) is the basic confidence level, pl (A) is the average confidence level, and U (A) is the uncertainty measure;
for an evidence item A, the basic trust level is calculated:
where BelT (a) represents the confidence in setting a under different evidence sources, this summation covering all evidence sources;
calculate the average confidence level Pl (a):
where ¬ A represents the complement of A and Bel (¬ A) represents the basic confidence that ¬ A is not believed;
uncertainty metric U (A) represents the strength of confidence in evidence:
calculate the resultant confidence level (CA):
where U (¬ A) represents the uncertainty measure of ¬ A and the resultant confidence level represents the final confidence level for setting A;
in STEP-3, the STEP of outputting the final control strategy:
(1) Selection of a threshold: in the composite confidence distribution vector, a threshold is selected to determine which states are considered trusted states and which states are not employed; the threshold value is selected to comprehensively consider the distribution of the trust degree and the requirement of the system;
(2) And (3) formulating a control strategy: according to the selected threshold value, the state exceeding the threshold value in the synthesized confidence level distribution is regarded as a trusted state, and the state not exceeding the threshold value is regarded as an untrusted state; based on this determination, different control strategies are formulated;
(3) The final control strategy of the execution control strategy is formulated according to the synthesized trust degree distribution, and the system executes corresponding operations according to the strategy, including adjusting the temperature control system of the building, and switching on or off the refrigeration equipment and the heating equipment to realize the selected state.
2. The building intelligent control method according to claim 1, wherein: in STEP-2.2, the markov property and initial state distribution is such that 50% of the probability is in the cooling mode, 30% of the probability is in the heating mode, and 20% of the probability is in the off mode;
the state distribution of the building system is:
the initial state distribution is inThe initial node of the Markov chain.
3. The building intelligent control method according to claim 2, wherein: in STEP-2.3, in the transition probability matrix P, P (i, j) represents the probability of transition from state i to state j:
the non-negativity: for all i and j, the following is satisfied:
The normalization: for each state i, the following is satisfied:
where Σ is the sum of all possible states j;
the transition probability is between 0 and 1:
for all i and j, the following is satisfied:
。
4. an intelligent building control system, which is characterized in that: the intelligent control system comprises a processor, and a memory coupled to the processor, wherein the memory stores program instructions that, when executed by the processor, cause the processor to perform the building intelligent control method of any of claims 1-3.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311465113.4A CN117193034B (en) | 2023-11-07 | 2023-11-07 | Building intelligent control method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311465113.4A CN117193034B (en) | 2023-11-07 | 2023-11-07 | Building intelligent control method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117193034A CN117193034A (en) | 2023-12-08 |
CN117193034B true CN117193034B (en) | 2024-02-02 |
Family
ID=88989080
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311465113.4A Active CN117193034B (en) | 2023-11-07 | 2023-11-07 | Building intelligent control method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117193034B (en) |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6400996B1 (en) * | 1999-02-01 | 2002-06-04 | Steven M. Hoffberg | Adaptive pattern recognition based control system and method |
JP2005251185A (en) * | 2004-02-05 | 2005-09-15 | Toenec Corp | Electric equipment diagnostic system |
CN102426674A (en) * | 2011-10-28 | 2012-04-25 | 山东电力集团公司青岛供电公司 | Power system load prediction method based on Markov chain |
JP2014178090A (en) * | 2013-03-15 | 2014-09-25 | Daikin Ind Ltd | Cooling and heating unit, air conditioning equipment, humidity control unit and humidity control device |
CN106461260A (en) * | 2014-05-15 | 2017-02-22 | 三星电子株式会社 | Method and apparatus for controlling temperature |
EP3267629A1 (en) * | 2016-07-05 | 2018-01-10 | Rheinisch-Westfälische Technische Hochschule (RWTH) Aachen | System for automated analysis and optimization of building automation systems |
EP3874390A1 (en) * | 2018-11-02 | 2021-09-08 | Arizona Board of Regents on behalf of the University of Arizona | Runtime adaptive risk assessment and automated mitigation |
CN113821903A (en) * | 2021-07-09 | 2021-12-21 | 腾讯科技(深圳)有限公司 | Temperature control method and device, modular data center and storage medium |
CN114017904A (en) * | 2021-11-04 | 2022-02-08 | 广东电网有限责任公司 | Operation control method and device for building HVAC system |
CN114498649A (en) * | 2022-01-04 | 2022-05-13 | 清华大学 | Active power distribution network building thermal load control method and device, electronic equipment and storage medium |
CN115907350A (en) * | 2022-11-04 | 2023-04-04 | 华北电力大学 | Energy management method and system of building comprehensive energy system |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8249731B2 (en) * | 2007-05-24 | 2012-08-21 | Alexander Bach Tran | Smart air ventilation system |
US10012406B2 (en) * | 2014-05-15 | 2018-07-03 | Samsung Electronics Co., Ltd. | Method and apparatus for controlling temperature |
US20210090188A1 (en) * | 2019-09-24 | 2021-03-25 | International Business Machines Corporation | Iot based property condition evaluation system |
-
2023
- 2023-11-07 CN CN202311465113.4A patent/CN117193034B/en active Active
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6400996B1 (en) * | 1999-02-01 | 2002-06-04 | Steven M. Hoffberg | Adaptive pattern recognition based control system and method |
JP2005251185A (en) * | 2004-02-05 | 2005-09-15 | Toenec Corp | Electric equipment diagnostic system |
CN102426674A (en) * | 2011-10-28 | 2012-04-25 | 山东电力集团公司青岛供电公司 | Power system load prediction method based on Markov chain |
JP2014178090A (en) * | 2013-03-15 | 2014-09-25 | Daikin Ind Ltd | Cooling and heating unit, air conditioning equipment, humidity control unit and humidity control device |
CN106461260A (en) * | 2014-05-15 | 2017-02-22 | 三星电子株式会社 | Method and apparatus for controlling temperature |
EP3267629A1 (en) * | 2016-07-05 | 2018-01-10 | Rheinisch-Westfälische Technische Hochschule (RWTH) Aachen | System for automated analysis and optimization of building automation systems |
EP3874390A1 (en) * | 2018-11-02 | 2021-09-08 | Arizona Board of Regents on behalf of the University of Arizona | Runtime adaptive risk assessment and automated mitigation |
CN113821903A (en) * | 2021-07-09 | 2021-12-21 | 腾讯科技(深圳)有限公司 | Temperature control method and device, modular data center and storage medium |
CN114017904A (en) * | 2021-11-04 | 2022-02-08 | 广东电网有限责任公司 | Operation control method and device for building HVAC system |
CN114498649A (en) * | 2022-01-04 | 2022-05-13 | 清华大学 | Active power distribution network building thermal load control method and device, electronic equipment and storage medium |
CN115907350A (en) * | 2022-11-04 | 2023-04-04 | 华北电力大学 | Energy management method and system of building comprehensive energy system |
Non-Patent Citations (2)
Title |
---|
一种信度马尔科夫模型及应用;邓鑫洋, 邓勇,章雅娟, 刘琪;自动化学报;全文 * |
一种基于马尔科夫链的冲突证据组合方法;李新德,董清泉, 王丰羽,雒超民;自动化学报;第1-13页 * |
Also Published As
Publication number | Publication date |
---|---|
CN117193034A (en) | 2023-12-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP2013669B1 (en) | A method of optimising energy consumption | |
Korkas et al. | Grid-connected microgrids: Demand management via distributed control and human-in-the-loop optimization | |
US10371405B2 (en) | Building power management systems | |
Sanandaji et al. | Fast regulation service provision via aggregation of thermostatically controlled loads | |
Hagras et al. | An intelligent agent based approach for energy management in commercial buildings | |
Ouf et al. | A simulation-based method to investigate occupant-centric controls | |
CN111664550A (en) | Energy efficiency optimization method and system based on prediction model and optimization algorithm | |
Baniyounes et al. | Institutional smart buildings energy audit. | |
CN116266253A (en) | Optimization control method, system and computer readable storage medium for air conditioner parameters | |
CN113110057A (en) | Heating power station energy-saving control method based on artificial intelligence and intelligent decision system | |
CN113110056B (en) | Heat supply intelligent decision-making method and intelligent decision-making machine based on artificial intelligence | |
Ku et al. | Building electric energy prediction modeling for BEMS using easily obtainable weather factors with Kriging model and data mining | |
CN117193034B (en) | Building intelligent control method and system | |
CN117270410A (en) | Smart home energy consumption control method, device and computer readable storage medium | |
Mpelogianni et al. | Proactive building energy management methods based on Fuzzy Logic and Expert Intelligence | |
CN116963461A (en) | Energy saving method and device for machine room air conditioner | |
Coffman et al. | A study of virtual energy storage from thermostatically controlled loads under time-varying weather conditions | |
CN112821456B (en) | Distributed source-storage-load matching method and device based on transfer learning | |
Harish et al. | Simulation based energy control and comfort management in buildings using multi-objective optimization routine | |
CN116187910B (en) | Temperature control warehouse capacity configuration method and device, electronic equipment and storage medium | |
KR102498340B1 (en) | Apparatus and method for determining an operation schedule of heat pumps based on building energy management system | |
Korkas et al. | Adaptive optimization for smart operation of cyber-physical systems: A thermostatic zoning test case | |
Yazdkhasti et al. | A methodology to forecast the control capacity of a population of thermostatically controlled appliances in a demand-side management | |
EP4361740A1 (en) | A computer-implemented method for configuring a controller for a technical system | |
CN117439101A (en) | Intelligent network for interaction of new energy and flexible load in power grid |
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 |