Search

CN-121977246-A - Winter heating thermal disturbance recognition and alliance game optimization method and system for open type office environment

CN121977246ACN 121977246 ACN121977246 ACN 121977246ACN-121977246-A

Abstract

The invention discloses a method and a system for identifying heating thermal disturbance in winter and optimizing alliance games in an open type office environment, and belongs to the technical field of intelligent control of artificial environments. Aiming at the problems of unknown use state of local heating equipment (Personal Comfort Systems, PCS for short) in open office environment winter heating, energy consumption waste, thermal comfort conflict, poor suitability of multi-unit configuration and the like caused by lack of cooperation of a single or multiple office space winter hot air heating units and the PCS, the invention collects temperature time sequence data of each micro-area through a distributed environment sensor, inverts PCS open state and thermal independent coefficient by utilizing a thermal disturbance recognition algorithm, clusters the micro-area into three types of virtual alliances of rich heat source, lean heat source and neutrality based on the coefficient, and constructs an alliance model which takes global set temperature of the single unit or joint set temperature of the multiple units as decision variables, fuses thermal comfort effect and global energy consumption penalty items by introducing Shapley values to evaluate marginal contribution and tolerance cost of each alliance and solve balanced solution or pareto optimal solution of pareto.

Inventors

  • CAI HAO
  • YU GUO
  • LI TAO
  • TAN MEILAN

Assignees

  • 南京工业大学

Dates

Publication Date
20260505
Application Date
20260107

Claims (12)

  1. 1. A winter heating thermal disturbance recognition and alliance game optimization method and system for an open type office environment are characterized by comprising the following steps: The method comprises the steps of S1, dividing an open type office environment space into a plurality of micro areas according to station layout, and collecting air temperature time sequence data of each micro area in winter heating through a sensor network, wherein winter heating equipment of the open type office environment space is one or more office space winter hot air heating units without primary independent area temperature regulation functions, the units cover conventional forms such as independent heat pump air conditioners, a centralized hot air system, a fan coil, a multi-connected indoor unit and the like, and the heating units are centralized control units which do not support independent area temperature regulation; Step S2, based on the collected temperature data and the first derivative temperature change rate dT i /dT, combining the current running state of the hot air heating unit in winter in each office space, executing feature extraction through an environmental thermal disturbance recognition algorithm, recognizing local environmental thermal disturbance existing in each micro-area, wherein the environmental thermal disturbance at least comprises a PCS heat source, a personnel heat source or an environmental radiation heat source, inverting the PCS starting state and calculating a thermal independent coefficient gamma i of each micro-area, wherein the thermal independent coefficient gamma i is used for representing the capability of the micro-area not influenced by global heating and relying on the local heat source to maintain thermal comfort; Step S3, dynamically classifying each micro-area into three types of virtual alliances with different game strategies according to the numerical distribution characteristics of the thermal independent coefficient gamma i ; Step S4, if a heating unit is used, constructing a coalition game model taking the global set temperature of the unit as a decision variable, if a plurality of heating units are used, dividing temperature acquisition subareas according to unit coverage areas, building association mapping of virtual coalitions and corresponding coverage units, and constructing a coalition game model taking the set temperatures of the plurality of units as the combined decision variable; And S5, introducing a shape value to evaluate marginal contribution and tolerance cost of each virtual alliance, generating a control instruction containing target temperature and wind speed, and driving the heating unit to be adjusted to the set state through an execution interface.
  2. 2. The method of claim 1, wherein the distinguishing and identifying logic for PCS heat sources from ambient interference in step S2 is: Calculating the temperature change rate dT i /dT of the micro-area i, which is physical sensing data and outputs an air conditioner control instruction, if the detected dT i /dT is larger than a preset disturbance threshold alpha in a time window in which the air supply heat of the central air conditioner is not remarkably increased, and the temperature rise is limited to the micro-area i and a peripheral minimum range (presenting a localized characteristic), judging that the micro-area i starts local heating equipment and sets a heat independent coefficient gamma i of the micro-area i to be a high value, if the micro-area temperature is not suddenly changed and is lower than the space average temperature for a long time, judging that the micro-area temperature is a low heat independent area, setting a heat independent coefficient gamma i to be a low value, and otherwise judging that the heat independent coefficient gamma i is a neutral heat independent area.
  3. 3. The method of claim 2, wherein the preset time window is a reasonable time length capable of identifying local temperature rise mutation, the time length ranges from 3 minutes to 12 minutes, and the preset disturbance threshold alpha is a reasonable threshold capable of distinguishing PCS heat sources from environmental interference, and the threshold range is 0.2 ℃ to 1.2 ℃ per minute.
  4. 4. A method according to claim 3, wherein the predetermined time window is preferably 5-10 minutes and the predetermined disturbance threshold α is preferably 0.3-1.0 ℃ per minute.
  5. 5. The method of claim 2, wherein the thermal independent coefficient γ i has a high value range of 0.7 or more, a low value range of 0.3 or less, a median value range of 0.3 to 0.7, preferably a high value range of 0.8 or more, a low value range of 0.2 or less, and a median value range of 0.2 to 0.8.
  6. 6. The method according to claim 1, wherein the virtual federation of step S3 includes: 1) The heat-rich source alliance consists of micro areas with heat independent coefficients gamma i larger than a preset value, and the game strategy is that the air conditioner set temperature tends to be lower so as to avoid discomfort caused by heat superposition; 2) The lean heat source alliance consists of micro-areas with the heat independent coefficient gamma i smaller than a preset value, and the game strategy is that the air conditioner set temperature is prone to be higher so as to meet basic thermal comfort. 3) The neutral alliance consists of a micro-area with a heat independent coefficient gamma i as a median value, and the game strategy is to accept the fluctuation of the set temperature in a conventional comfort interval without strong adjustment requirements.
  7. 7. The method of claim 6, wherein the set temperature of the heating unit with the tendency of the heat-rich source alliance is less than or equal to 22 ℃, the set temperature of the heating unit with the tendency of the heat-poor source alliance is more than or equal to 23 ℃, the fluctuation range of the set temperature accepted by the neutral alliance is 21-23 ℃, preferably, the set temperature of the heat-rich source alliance is 20-22 ℃, the set temperature of the heat-poor source alliance is 23-25 ℃, and the fluctuation range of the neutral alliance is 21.5-22.5 ℃.
  8. 8. The method of claim 1, wherein the integrated utility function of the gaming model in step S4 is u= Σ (ω k C k ) - λe, wherein: Omega k is the weight coefficient of the k-th type virtual alliance, and is dynamically distributed based on the heat independent coefficient, wherein the higher the heat independent coefficient is, the lower the weight ratio is; c k is the thermal comfort effect of the k-th virtual alliance, and is calculated based on the deviation between the actual temperature and the comfort temperature of each micro area (1-5 minutes quantization); E is global total energy consumption (kWh) of one or more office space winter hot air type heating units, and lambda is an energy consumption penalty coefficient.
  9. 9. The method of claim 8, wherein the weight coefficients are distributed in a range of 8% -20% of the heat-rich source alliance weight, 35% -65% of the neutral alliance weight and 20% -45% of the lean alliance weight, the energy consumption penalty coefficient lambda is 0.1% -0.5, preferably 10% -15% of the heat-rich source alliance weight, 40% -60% of the neutral alliance weight, 25% -40% of the lean alliance weight, and the energy consumption penalty coefficient lambda is 0.2% -0.4.
  10. 10. The method of claim 1, wherein in a scenario of a plurality of hot air heating units in office space in winter, association mapping rules of the alliances and the units are that each virtual alliance is bound to a covering unit corresponding to a temperature acquisition subarea where the virtual alliance is located, and if a single alliance spans a plurality of subareas, the virtual alliance is bound to a covering unit corresponding to a main distribution area of the alliance, or distributed to the corresponding units according to distribution proportion to participate in a joint decision.
  11. 11. An open office environment winter heating thermal disturbance recognition and alliance game optimization system for implementing the method of any one of claims 1-10, comprising: 1) The sensing module consists of wireless temperature and humidity sensors distributed in each micro area of the open office environment space, wherein the sampling period of the sensors is reasonable duration capable of capturing temperature abrupt change, and the data transmission delay is less than or equal to 3 seconds, and is used for acquiring air temperature time sequence data of each micro area during heating in winter; 2) The data processing module is used for receiving the data acquired by the sensing module, executing temperature change rate calculation and an environmental thermal disturbance recognition algorithm, inverting the PCS starting state and outputting the thermal independent coefficient of each micro area; 3) The game controller is internally provided with a game theory algorithm engine and is used for dynamically constructing a virtual alliance according to the thermal independent coefficient, generating a comprehensive utility function, respectively solving a Nash equilibrium solution or a Pareto optimal solution aiming at one or more machine group scenes, and outputting the optimal set temperature of the hot air heating machine group in winter corresponding to the office space; 4) The execution interface is used for communicating with one or more winter hot air heating units (including independent heat pump air conditioner, centralized hot air system, fan coil, multi-split indoor unit and the like) in the open office environment space, adapting to conventional control protocols such as RS485, infrared or Bluetooth and the like and issuing temperature and air quantity setting instructions.
  12. 12. The system according to claim 11, wherein the sensor sampling period of the sensing module is 0.5-3 minutes, preferably 1-2 minutes.

Description

Winter heating thermal disturbance recognition and alliance game optimization method and system for open type office environment Technical Field The invention relates to the technical field of building environments and equipment engineering, in particular to a thermal disturbance recognition and alliance game optimization system and method which are suitable for an open office environment winter heating scene, coordinate one or more office space winter hot air heating units and distributed local heating equipment (PCS) operation through a game theory by utilizing low-cost environment sensing data. Background In winter heating control in open office environment space (without physical partitions, centralized layout of stations, mutual influence of thermal environments in each area) is a serious challenge. Whether single or multiple heating units are configured, the method has the remarkable technical problems that: after comparing with prior art search reports and related documents, it was found that the prior art has mainly the following problems: 1. The existing game control model ignores the unstable state thermal disturbance, so that control mismatch is caused, the existing HVAC game control method generally assumes that the environment is in a quasi-stable state, and the unstable state thermal disturbance frequently occurring in an office space is ignored. For example, cold air intrusion caused by a user opening and closing a door, instantaneous turning on or off of a local heating facility (PCS). The prior art lacks a front-end thermal disturbance characteristic extraction module, and cannot distinguish 'systematic load change' and 'local instantaneous disturbance', so that a game model misjudges a local PCS heat source as environment heating to reduce unit output or delay instantaneous cold impact reaction caused by opening and closing a door. The application introduces thermal disturbance recognition to solve the 'perception-decision' fault problem. 2. The heat coupling is strong, the demand conflict is remarkable, the open space is free from physical separation, the local temperature difference reaches 3-5 ℃ due to factors such as cold air permeation at the window edge, personnel accumulation in the inner area, personnel flow at the door, and the like, the physical difference of superimposed personnel (the personnel who feel cold start PCS), the mode of 'global unified control' of a single unit or 'independent control of a plurality of units without cooperation' is difficult to simultaneously meet the differential heat demands of different areas, the prior art is only regulated based on a single target of temperature deviation, or quantitative heat demand classification basis is lacked, and the conflict cannot be effectively coordinated. The energy waste is caused by 'heat superposition', the traditional heating unit control system cannot sense a 'hidden heat source' formed by starting PCS, a single unit continuously supplies heat in full power, a plurality of units are respectively used for excessively heating, the problem of local overheating is further amplified by the heat diffusion characteristic of an open space, the heating Coefficient (COP) of the heating unit is obviously reduced due to a low-temperature high-humidity environment in winter, the energy waste is more prominent than other areas, the hidden heat source is not identified in the prior art, or closed loop logic of 'identification-coordination-optimization' is not formed only through hardware linkage control. 4. The improved threshold is high, the compatibility is poor, the existing intelligent control scheme generally requires that all PCS are connected to the Internet of things or are replaced by an expensive VAV (variable air volume) system, the implementation cost is high, office layout can be damaged, the intelligent control scheme is not suitable for a large number of existing open office environment buildings, meanwhile, the existing scheme does not consider the configuration difference of a single unit/a plurality of conventional heating units, the cooperative control requirement of a plurality of units can not be met, and a soft cooperative scheme is not provided for the conventional units without the primary independent area temperature regulation function, so that the universality is insufficient. 5. The prior art is not provided with a multi-objective cooperative game mechanism, namely, only a single objective (such as thermal comfort or energy consumption) is concerned, or a simple priority distribution is adopted, a multi-area thermal demand game coordination model is not constructed, a fair quantization benefit distribution mechanism (such as Shapley value) is not introduced, marginal contribution and tolerance cost of each area cannot be balanced, and the fairness and effectiveness of an optimization result are insufficient. Disclosure of Invention The invention aims at providing a low-cost, high-compatibility and m