Smart hotel guest room intelligent regulation and control method and system based on multi-source data

Abstract

The invention discloses a smart hotel guest room intelligent regulation and control method and system based on multi-source data, and relates to the technical field of smart hotel guest room environment regulation and control and building energy consumption management, and the method comprises the steps: receiving a power grid demand response signal through a central controller, and obtaining the environment parameters, occupancy states and equipment operation information of each guest room; an upper-layer collaborative model with the aim of minimizing the overall energy consumption cost is constructed, and energy consumption regulation and control parameters of all guest rooms are decomposed and generated and issued to a guest room local controller; and the guest room local controller maps the regulation and control parameters into an energy consumption constraint and a cost coefficient, and based on a preset non-cooperative game model, an environment regulation and control equipment optimal control instruction meeting the guest comfort constraint is obtained and executed. According to the method, deep fusion of power grid demands and guest room multi-source data is realized, personalized comfortable experience is taken into account while the demand response peak clipping target is completed, the comprehensive energy consumption cost of hotels is reduced, and the flexibility and refinement level of energy regulation and control is improved.

Description

Technical Field

[0001] This invention relates to the field of smart hotel room environmental control and building energy consumption management technology, specifically a smart hotel room intelligent control method and system based on multi-source data. Background Technology

[0002] With the development of technologies such as the Internet of Things (IoT), building automation, and cloud computing, smart hotel room management systems have gradually evolved from traditional on / off control and timetable control to a comprehensive platform integrating guest room control systems (GRMS), building energy management systems (BEMS), and hotel business management systems. By deploying sensors in guest rooms to monitor temperature, humidity, carbon dioxide concentration, illuminance, and human presence, and combining these with business systems such as door lock systems, reservation systems, and membership management systems, the system can perceive the environmental status, occupancy status, and some user preferences in the guest rooms. Based on this, hotels are beginning to experiment with energy-saving strategies, scenario-based modes, load-level control, and simple demand response strategies to reduce the energy consumption of environmental control equipment such as air conditioning and fresh air systems, improve the precision of operation and management, and begin preliminary collaboration with peak-valley electricity pricing and demand response projects on the power side.

[0003] However, existing smart hotel room control technologies are generally still based on localized adjustments for individual rooms or global energy-saving control based on empirical rules. This makes it difficult to achieve an optimal balance between overall hotel energy costs and guest comfort across multiple rooms under complex constraints. On one hand, while multi-source data can be collected, it is mostly limited to display, alarms, or simple threshold control, lacking the ability to uniformly model and collaboratively optimize grid demand response signals, room operating status parameters, historical energy consumption data, and user preference data. Central control systems typically employ fixed temperature adjustment ranges and uniform power limiting strategies, failing to differentiate control based on the real-time status and preferences of different rooms. On the other hand, existing demand response solutions often focus on the grid or the hotel as a whole, only considering total load reduction and time windows. They lack quantitative characterization and constraints on changes in individual room comfort, and cannot model and weigh the potential compensation costs resulting from decreased comfort at the cost function level. Meanwhile, the local control logic of guest rooms usually only uses simple PID, fuzzy control or static scene switching, without establishing a unified decision-making framework for the overall energy consumption constraints of the hotel and the utility of local guests through non-cooperative game theory, resulting in a lack of effective coupling between central collaborative control and local personalized control. It is difficult to achieve cross-guest room energy consumption coordination and dynamic balance of guest comfort under the constraints of grid demand response, and thus cannot achieve the effect of the two-layer collaborative optimization technology that this invention aims to achieve. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by this invention is that existing intelligent control methods for smart hotel rooms cannot uniformly model and finely perceive the operating status of rooms and user needs based on multi-source data, cannot coordinate with the power grid demand response in a timely manner to optimize the overall energy consumption cost of the hotel, cannot take into account the personalized comfort of guests in different rooms under global energy consumption constraints, and cannot construct an intelligent method that combines central collaborative optimization with local non-cooperative game theory in guest rooms to achieve coordinated control of overall hotel energy consumption and guest room comfort under demand response constraints.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, embodiments of the present invention provide a smart hotel room intelligent control method based on multi-source data, applied to a system including a central controller and multiple local controllers for guest rooms communicatively connected to the central controller, characterized by comprising the following steps: The central controller receives power grid demand response signals and acquires the operating status parameters of each guest room; Based on the demand response signal and the operating status parameters, the central controller constructs an upper-level collaborative model with the goal of minimizing overall energy consumption costs, solves and generates energy consumption control parameters for each guest room, and distributes the energy consumption control parameters to the local controllers of each guest room respectively. The guest room local controller receives the energy consumption control parameters for the guest room and adjusts its internal preset non-cooperative game model based on the energy consumption control parameters. The guest room local controller, based on the adjusted non-cooperative game model, solves for and executes the optimal control commands for the environmental control equipment in the guest room.

[0007] As a preferred embodiment of the intelligent hotel room control method based on multi-source data described in this invention, wherein the central controller receives the power grid demand response signal including: The central controller receives demand response signals periodically or when a demand response event is triggered through a communication connection established with the grid-side dispatch platform or demand response platform; the demand response signal includes at least the target load adjustment amount, the demand response execution time window, and incentive price parameters.

[0008] As a preferred embodiment of the intelligent control method for smart hotel rooms based on multi-source data described in this invention, the operating status parameters include: Temperature, humidity, carbon dioxide concentration, and illuminance parameters were collected by environmental sensors in the guest rooms; Guest room occupancy status information obtained from guest room human body sensors and door lock systems; The operating mode, set parameters, and real-time power reported by the guest room environment control equipment; Historical energy consumption data and user preference data associated with guest rooms.

[0009] As a preferred embodiment of the intelligent hotel room control method based on multi-source data described in this invention, the construction of the upper-level collaborative model with the goal of minimizing overall energy consumption cost includes: Based on the target load adjustment amount, demand response execution time window, and incentive price parameter in the demand response signal, an upper-level collaborative model is constructed with the objective function of minimizing the hotel's total electricity cost within the time window; wherein, the total electricity cost includes the energy consumption cost determined based on the incentive price parameter and the load adjustment amount of each guest room, as well as the compensation cost caused by the load adjustment leading to the indoor environment deviating from the baseline comfort level. The operating status parameters are used as constraints, which include at least the following: the sum of the load adjustment amounts of each guest room meets the target load adjustment amount, and the load adjustment amount of each guest room is within its corresponding allowable operating range.

[0010] As a preferred embodiment of the intelligent control method for smart hotel rooms based on multi-source data described in this invention, the step of solving and generating energy consumption control parameters corresponding to each room includes: A distributed optimization algorithm is used to solve the upper-level collaborative model to obtain the adjustable energy consumption budget allocated to each guest room within the demand response execution time window; The adjustable energy consumption budget is combined with the incentive price parameter to generate energy consumption control parameters for each guest room, which are then sent to the corresponding guest room local controller via the hotel's internal communication network.

[0011] As a preferred embodiment of the intelligent control method for smart hotel rooms based on multi-source data described in this invention, the step of adjusting the internally preset non-cooperative game model based on the energy consumption control parameters includes: The adjustable energy consumption budget is used as the energy consumption constraint parameter in the non-cooperative game model, and the incentive price parameter is used as the energy consumption cost coefficient in the non-cooperative game model.

[0012] As a preferred embodiment of the intelligent control method for smart hotel rooms based on multi-source data described in this invention, the non-cooperative game model includes: Guest comfort utility function and room energy cost function; The guest comfort utility function is a function of the deviation between the actual indoor environmental parameters and the user preference settings, and the guest room energy cost function is a function of the actual power of the environmental control equipment and the incentive price parameter.

[0013] As a preferred embodiment of the intelligent control method for smart hotel rooms based on multi-source data described in this invention, the optimal control command for obtaining the environmental control equipment in this room includes: The predicted indoor environmental parameters determined by the operating settings of the environmental control equipment are within the comfort range determined based on the user preference data; the guest room local controller obtains the optimal operating settings of the environmental control equipment by solving the Nash equilibrium of the non-cooperative game model under the constraints, and generates corresponding equipment control commands.

[0014] As a preferred embodiment of the intelligent control method for smart hotel rooms based on multi-source data described in this invention, the device control command includes at least one of the following parameters: temperature setpoint of the air conditioner, operating mode, fan speed level, and fresh air valve opening.

[0015] Secondly, embodiments of the present invention provide a smart hotel room intelligent control system based on multi-source data, including: a central controller and multiple local room controllers that are communicatively connected to the central controller; The central controller includes: Signal receiving module: configured to receive demand response signals from the power grid side; Data aggregation module: configured to acquire the operating status parameters of each guest room; Collaborative optimization module: configured to construct an upper-level collaborative model with the goal of minimizing overall energy consumption cost based on the demand response signal and the operating status parameters, and solve the model to generate energy consumption control parameters for each guest room; Parameter distribution module: configured to distribute the energy consumption control parameters for each guest room generated by the solution to the corresponding local controller of the guest room; The guest room local controller includes: Parameter receiving module: configured to receive energy consumption control parameters from the central controller; Local decision-making module: It has a preset non-cooperative game model and is configured to adjust the non-cooperative game model based on the energy consumption control parameters, and solve for the optimal control command of the environmental control equipment in this guest room; Control execution module: configured to execute the optimal control command to control the environmental control equipment.

[0016] The beneficial effects of this invention are as follows: By constructing a two-layer optimization architecture of central coordination and local game theory, this invention deeply integrates power grid demand response signals with multi-source guest room data, thereby significantly reducing the overall energy consumption cost of the hotel and accurately responding to power grid dispatch while ensuring the personalized comfort of guests. At the same time, through a dynamic game theory mechanism, load adjustment targets are fairly allocated, making energy-saving control imperceptible to guests. Thus, it takes into account multiple objectives such as user experience, hotel operating efficiency, and flexible interaction of the energy system. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein: Figure 1 This is an overall flowchart of a smart hotel room intelligent control method based on multi-source data, provided in the first embodiment of the present invention. Figure 2 This is a device connection diagram for a smart hotel room intelligent control method based on multi-source data, provided in the first embodiment of the present invention. Figure 3 This is a structural diagram of the central controller of a smart hotel room intelligent control system based on multi-source data, provided in the first embodiment of the present invention. Figure 4 The first embodiment of the present invention provides a structural diagram of a local controller for a smart hotel room control system based on multi-source data. Detailed Implementation

[0018] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0019] Example 1, referring to Figure 1 As an embodiment of the present invention, a smart hotel room intelligent control method based on multi-source data is provided, which is applied to a system including a central controller and multiple local controllers of guest rooms that are communicatively connected to the central controller.

[0020] like Figure 2As shown, the system in this embodiment is deployed using a "cloud-edge-device" collaborative architecture. The central controller 200 is connected to multiple guest room local controllers 300-1 to 300-N through the hotel's internal network. Each guest room local controller interacts with environmental sensors, occupancy sensing devices, smart door locks, and environmental control devices (such as air conditioning terminals, fresh air equipment, lighting / curtains, etc.) located in the guest rooms.

[0021] The central controller can be deployed on an industrial server in the hotel's server room or on a cloud server, communicating with local controllers in each guest room via the hotel's internal network. In this embodiment, the central controller is preferably an industrial-grade server running a general-purpose operating system, used to run the hotel's intelligent energy control core software and data storage services. The guest room local controller is preferably a multi-protocol intelligent IoT gateway deployed in the low-voltage electrical box of each guest room, capable of interacting with environmental sensors, occupancy sensing devices, and environmental control equipment, and possessing certain local computing and storage resources to handle data aggregation and local intelligent decision-making for that guest room.

[0022] S1: The central controller receives the power grid demand response signal and obtains the operating status parameters of each guest room.

[0023] The central controller receives demand response signals periodically or when a demand response event is triggered through a communication connection established with the grid-side dispatch platform or demand response platform; the demand response signal includes at least the target load adjustment amount, the demand response execution time window, and incentive price parameters.

[0024] It should be noted that the central controller receives demand response signals periodically or when demand response events are triggered through a communication connection established with the grid-side dispatching platform or demand response platform. This communication connection can be built on a TCP / IP network and can use HTTP, HTTPS, message queues, or open demand response protocols in the power industry for data exchange. In an optional implementation, the central controller can integrate a client program conforming to the demand response standard protocol to maintain communication with the grid company or a third-party aggregator's demand response server through a secure and encrypted connection.

[0025] In practical implementation, the demand response signal can be sent in the form of a structured data message. The central controller parses the message and extracts at least the following information: Target load adjustment amount is used to indicate the target load that the hotel as a whole needs to reduce or adjust within the demand response execution time window; The demand response execution time window is used to indicate the start and end times of the demand response. Incentive price parameters are used to indicate the incentive price or compensation price corresponding to the adjustment per unit load during demand response execution.

[0026] In some scenarios, the target load adjustment can be directly provided by the grid side; in other scenarios, the central controller can calculate the target load adjustment based on the difference between the historical baseline load and the target load. The parsed demand response signal can be timestamped and stored in the central controller's local database for subsequent construction of upper-level collaborative models.

[0027] To construct complete operational status parameters, in this embodiment, the central controller collects and integrates multiple types of source data through the guest room local controller, specifically including: Temperature, humidity, carbon dioxide concentration, and illuminance parameters are collected by environmental sensors in the guest rooms. Specifically, environmental sensors are installed in each guest room to collect these parameters. These sensors can be digital temperature and humidity sensors, carbon dioxide sensors based on non-dispersive infrared principles, or light sensors, and are mounted on walls or ceilings near typical activity areas. The sensors can connect to the guest room's local controller via wired bus (e.g., I²C, RS485) or short-range wireless communication (e.g., ZigBee, Bluetooth Low Energy). The guest room's local controller acquires these environmental parameters at preset intervals (e.g., 1-5 minutes), packages them with the guest room identifier and timestamp, and uploads them to the central controller. The central controller performs necessary filtering, outlier removal, and time alignment on the received environmental data to create a description of the current environmental status of each guest room. Guest room occupancy status information is obtained from guest room human body sensors and door lock systems. Specifically, occupancy sensing devices, such as infrared human body sensors and millimeter-wave radar presence sensors, are installed in guest rooms to detect human activity. Simultaneously, the smart door lock system reports unlocking, locking, and deadbolt events to the guest room local controller or the hotel management system. The guest room local controller can have a built-in lightweight state machine that combines occupancy sensor signals with door lock events for logical judgment. For example, if a person is detected with minimal movement for a period of time after check-in, it is determined to be in an "occupied – active" state; if only a stationary signal is detected for an extended period and the door is deadbolted, it is determined to be in an "occupied – resting" state; if no human presence is detected for an extended period after the door is unlocked, it is determined to be in an "unoccupied" or "vacant" state. The obtained occupancy status information is reported to the central controller at regular intervals or when the status changes, serving as part of the operating status parameters for determining the permissible operating range and the depth of demand response for each guest room. The system receives operating modes, set parameters, and real-time power data reported by the guest room environmental control equipment. Specifically, the environmental control equipment in the guest rooms includes air conditioning terminals, fresh air systems, and optional smart lighting and curtains. Air conditioning equipment can be networked indoor units supporting standard communication protocols (such as Modbus and BACnet) or air conditioning terminals connected via a dedicated gateway. Fresh air systems can control the fresh air volume through electrically adjustable dampers or variable air volume (VAV) terminals. The local guest room controller reads or receives the operating modes (cooling, heating, ventilation, etc.), current set parameters (such as set temperature and fan speed), and real-time or estimated power values ​​from the environmental control equipment according to a preset sampling period. This data is then reported to the central controller in structured data format, enabling the central controller to understand the current operating status and load level of the equipment in each guest room. Historical energy consumption data and user preference data associated with guest rooms are also available. Specifically, the central controller can interact with the hotel's energy management system, property management system, and membership management system to read historical energy consumption data and user preference data associated with each guest room. Historical energy consumption data can include time-of-use electricity consumption, typical operating modes, and peak power for each guest room over a period of time (e.g., the past week or month), used to characterize the long-term energy consumption features of that room. User preference data can include the target temperature range set by guests during previous stays, preferred comfort scenarios (e.g., "cooler," "warmer," "quieter priority"), and comfort parameter templates preset by the hotel for different customer groups (business travelers, families, long-stay guests, etc.). For guest rooms with first-time occupancy and lacking historical data, the central controller can select default preference data corresponding to the guest type or room type and gradually adjust it during subsequent stays.

[0028] Furthermore, after receiving environmental data, occupancy status information, equipment operating parameters, historical energy consumption data, and user preference data from local controllers in each guest room and related business systems, the central controller preprocesses and integrates the aforementioned multi-source data. Specifically, this includes: Time alignment is performed on various types of data using a uniform time step, and data with sampling time deviations within a preset threshold are interpolated or aligned. For sensor readings that are clearly outside the reasonable range, perform anomaly detection and filtering; for short-term missing data, interpolate or replace with the most recent valid value. Using the guest room identifier as an index, the environmental parameters, occupancy status, equipment operating parameters, historical energy consumption statistics, and user preference parameters of the same guest room at the current moment are associated to construct the corresponding operating status parameter vector of the guest room.

[0029] Through the above steps, the central controller can continuously acquire and update the operating status parameters that reflect the real-time operating conditions and long-term characteristics of each guest room before the start of the demand response execution time window and during the execution process, providing a complete and reliable data foundation for the subsequent construction of an upper-level collaborative model with the goal of minimizing overall energy consumption costs.

[0030] S2: Based on the demand response signal and the operating status parameters, the central controller constructs an upper-level collaborative model with the goal of minimizing overall energy consumption costs, solves and generates energy consumption control parameters for each guest room, and sends the energy consumption control parameters to the local controllers of each guest room respectively.

[0031] Based on the target load adjustment amount, demand response execution time window, and incentive price parameter in the demand response signal, an upper-level collaborative model is constructed with the objective function of minimizing the hotel's total electricity cost within the time window; wherein, the total electricity cost includes the energy consumption cost determined based on the incentive price parameter and the load adjustment amount of each guest room, as well as the compensation cost caused by the load adjustment leading to the indoor environment deviating from the baseline comfort level. The operating status parameters are used as constraints, which include at least the following: the sum of the load adjustment amounts of each guest room meets the target load adjustment amount, and the load adjustment amount of each guest room is within its corresponding allowable operating range.

[0032] A distributed optimization algorithm is used to solve the upper-level collaborative model to obtain the adjustable energy consumption budget allocated to each guest room within the demand response execution time window; The adjustable energy consumption budget is combined with the incentive price parameter to generate energy consumption control parameters for each guest room, which are then sent to the corresponding guest room local controller via the hotel's internal communication network.

[0033] Specifically, after completing the data preparation in step S1, the central controller executes step S2. Its core task is to construct a global optimization model to solve for the load allocation scheme of each guest room that minimizes the overall energy consumption cost of the hotel while meeting the grid demand response target, and to convert this scheme into control commands that can be issued.

[0034] The central controller first divides the duration of the demand response into multiple consecutive, equally long control cycles (e.g., each cycle is 15 minutes). The decision objective of the upper-level collaborative model is to determine a reasonable load adjustment (i.e., the increase or decrease in load relative to the load required for normal operation) for each guest room within each control cycle.

[0035] The model aims to minimize the total electricity cost, which is designed as the sum of "economic expenditure" and "comfort compensation" to achieve an automatic trade-off. Energy Costs (Economic Expenditure): This part directly corresponds to the economic incentives from the power grid. The calculation principle is to multiply the unit incentive price provided by the power grid for this response event by the load adjustment amount undertaken by each guest room in each cycle, and then sum the results for all guest rooms and all cycles. It directly represents the total economic compensation the hotel can obtain by adjusting its own electricity load (if it's a load reduction) or the additional costs it needs to pay (if it's an increased load).

[0036] Comfort Compensation Cost (Experience Quantification): This part quantifies the potential impact of load adjustments on guest comfort in monetary terms. The central controller predicts how indoor environmental parameters (primarily temperature) will change when the load is adjusted, based on the physical characteristics of each guest room (e.g., room size, orientation, insulation performance) and current operating status. This predicted change is then compared to a baseline comfort range derived from user preference data. The greater the deviation, the higher the calculated compensation cost. By incorporating this cost, the model automatically avoids strategies that could significantly cause guest discomfort while maximizing economic benefits.

[0037] Solving the model must adhere to two types of hard constraints: Global task constraint (meeting grid requirements): In all control cycles, the sum of the load adjustments for all guest rooms must reach or be very close to the target load adjustment amount issued by the grid. This is the bottom line to ensure that the hotel as a whole completes its demand response task.

[0038] Individual capacity constraints (respecting room status): The load adjustment for each room in each cycle must not exceed its dynamic operating allowable range. This range is determined by the central controller based on the real-time status of the room. For guest rooms in an "occupied / active" state, in order to ensure the guest experience, the adjustment range is narrow, usually only allowing adjustments within a small range to ensure that the indoor temperature is always maintained near the user's preferred range; For "vacant" or "unoccupied" guest rooms, the scope of adjustment is greatly expanded, allowing for significant load reduction or increase within the limits of safe equipment operation, so that it can undertake the main response tasks. This range is obtained by inversely extrapolating the adjustable power range of equipment such as air conditioners from the permissible temperature range and the thermodynamic model of the guest room.

[0039] Furthermore, due to the large number of guest rooms, using a single centralized solution method may result in high computational complexity and insufficient real-time performance. Therefore, in this embodiment, the central controller uses a distributed optimization algorithm to solve the upper-layer collaborative model, achieving efficient parallel computing by decomposing a large-scale global problem into multiple structurally similar local subproblems.

[0040] Distributed collaborative solution process: In a preferred embodiment, the central controller internally decomposes the global optimization problem into multiple local subproblems along the room dimension, and iteratively solves them using a coordination mechanism of distributed optimization algorithms (such as the alternating direction multiplier method or dual decomposition framework). This process can be summarized as follows: Problem decomposition and initialization: The central controller assigns initial coordination parameters (such as global guidance prices or Lagrange multipliers) to the local subproblems corresponding to each guest room based on the overall goal set by the power grid.

[0041] Parallel local computation: The central controller solves each local subproblem in parallel internally. Under the constraints of given coordination parameters and its own operational limits, each subproblem independently calculates an optimal and feasible local load adjustment scheme for the current guest rooms and its estimated cost.

[0042] Global Coordination and Feedback: The central controller aggregates all local calculation schemes and checks the deviation between their sum and the target load adjustment. If the deviation exceeds a preset threshold, the coordination parameters are updated (e.g., adjusting the global guidance price) according to the rules of the distributed optimization algorithm, and this is fed back into the next round of local calculations.

[0043] Iteration to convergence: Repeatedly execute parallel local computation and global coordination steps. After several rounds of iteration, each local solution is continuously modified under the guidance of coordination parameters, eventually ensuring that the sum of all local solutions precisely meets the grid requirements, while the overall cost tends to stabilize, reaching a globally optimal or approximate equilibrium solution.

[0044] Through the above methods, the central controller independently completes the construction and solution of the upper-level collaborative model, generating load adjustment schemes for each guest room. In other implementations, the local controllers of the guest rooms can also undertake some of the computational tasks to enhance privacy, but the core global coordination and convergence determination are still the responsibility of the central controller.

[0045] Generation of energy consumption control parameters: Once the solution is obtained, the central controller generates a core instruction for each guest room: an adjustable energy budget. This budget is not a hard power limit for every moment, but rather a total, flexible energy adjustment allowance allocated to the guest room throughout the entire demand response period (e.g., "Your total electricity consumption can be reduced by 3 kWh compared to the baseline in the next two hours").

[0046] Subsequently, the central controller binds the adjustable energy consumption budget of each guest room to the incentive price parameters of this event, packages them into structured energy consumption control parameters, and distributes them to the corresponding guest room local controllers through the hotel's internal network.

[0047] Thus, step S2 successfully transformed the macroscopic, rigid power grid dispatch instructions into a series of microscopic, flexible "budget-price" combination instructions. These instructions defined the action boundaries for each client and provided economic incentives, laying the foundation for the local controller to make autonomous and personalized game-theoretic decisions in step S3.

[0048] S3: The guest room local controller receives the energy consumption control parameters for the guest room and adjusts its internal preset non-cooperative game model based on the energy consumption control parameters.

[0049] The adjustable energy consumption budget is used as the energy consumption constraint parameter in the non-cooperative game model, and the incentive price parameter is used as the energy consumption cost coefficient in the non-cooperative game model.

[0050] The non-cooperative game model includes: a guest comfort utility function and a room energy cost function; the guest comfort utility function is a function of the deviation between the actual indoor environmental parameters and the user preference settings, and the room energy cost function is a function of the actual power of the environmental control equipment and the incentive price parameters.

[0051] Specifically, it should be noted that step S3 is initiated after the local controller in the guest room receives the energy consumption control parameters for that guest room from the central controller. Its core task is to parse and utilize the "budget-price" instructions issued from the upper level, adjust its internal preset non-cooperative game model, align the local decision-making objective with the overall optimization objective of the hotel, and prepare for the generation of the final equipment control instructions.

[0052] The guest room local controller receives structured data packets from the central controller via the hotel's internal communication network. These data packets correspond to the energy consumption control parameters generated for the guest room in step S2. The local controller parses these parameters and extracts at least the following two core elements: Adjustable Energy Budget: This indicates the amount of energy consumption adjustment allowed for this room relative to the baseline operating curve within the entire demand response execution window. This budget can be presented as a total indicator, such as "total electricity consumption can be reduced by up to 2.5 kWh compared to the normal operating baseline in the next two hours," or it can be refined into a sequence of upper / lower limits for power adjustments divided by control cycles. This budget constitutes a flexible total constraint or time-of-use power constraint on the energy consumption of local equipment.

[0053] Incentive price parameter: This parameter characterizes the economic incentive provided by the power grid for unit load adjustment in this demand response. It is usually given as a compensation price per unit of energy or power (e.g., a certain amount of compensation is obtained for every 1 kilowatt-hour saved). This parameter directly participates in the calculation of local energy consumption costs and is the "price signal" in local game theory.

[0054] After completing the data packet parsing, the guest room local controller stores the aforementioned budget and price parameters in the local configuration area and aligns them with the current demand response execution time window to ensure that subsequent decision calculations are performed on the correct time scale.

[0055] In this embodiment, each guest room's local controller has a pre-set non-cooperative game model to characterize and coordinate the interests between the "guest side seeking maximum comfort" and the "hotel management side seeking minimum energy consumption." Upon receiving energy consumption control parameters, the local controller parameterizes the game model as follows: Internalize the adjustable energy consumption budget into energy consumption constraint parameters; Based on the length of the demand response time window and the room's baseline load forecast, the local controller transforms the adjustable energy consumption budget issued by the upper layer into energy consumption constraints in the game theory model. During model solving, for any candidate control scheme (such as the air conditioning set temperature and fan speed within a series of control cycles), the local controller predicts the corresponding energy consumption changes and requires that the total energy consumption adjustment within all control cycles not exceed the budget limit. If the budget is given in the form of time-sharing power boundaries, then the power adjustment within each control cycle must fall within the corresponding upper / lower limit range. In this way, the adjustable energy consumption budget is explicitly written as an "energy consumption constraint parameter" in the local game theory model, ensuring that local decisions do not exceed the resource boundaries allocated to this room by the upper-level collaborative model. Set the incentive price parameter to the energy consumption cost coefficient; Simultaneously, the local controller incorporates the incentive price parameter into the energy cost calculation formula within the game theory model. For any candidate operating scheme, when calculating energy costs based on predicted power or energy usage, the local controller uses the incentive price parameter as a cost coefficient per unit of energy consumption for weighting: the higher the incentive price, the stronger the model's penalty for additional energy consumption, and the more significant the "reward" for energy-saving behavior. In this way, the incentive price parameter becomes a key coefficient in the guest room's energy cost function within the game theory model, thereby directly transmitting the economic incentives from the grid side into local decision-making.

[0056] Through the above two adjustments, the "budget-price" information output by the upper-level collaborative optimization is accurately embedded into the local non-cooperative game model. The macro-level resource allocation results are transformed into locally executable constraints and cost weights, realizing the effective transmission of global objectives to local decision-making rules.

[0057] Furthermore, after the adjustment, the local non-cooperative game model essentially consists of two interacting utility / cost functions, corresponding to two "virtual participants" in the game: Guest comfort utility function: This function simulates guests' comfort levels under different environmental control schemes. Its input is the deviation between actual or predicted indoor environmental parameters (such as temperature and humidity) and set values ​​or comfort ranges obtained from user preference data. The output is a utility value representing the comfort level. A smaller deviation results in a higher utility value; a larger deviation results in a lower utility value. Essentially, this function is "a function of the deviation between actual indoor environmental parameters and user preference set values."

[0058] Guest room energy cost function: This function characterizes the economic cost of guest room energy consumption under a specific operating scheme. Its inputs mainly include the actual power or energy usage of the environmental control equipment in each control cycle, and the corresponding incentive price parameters. The output is the total energy cost of the operating scheme. The higher the power and the higher the incentive price, the larger the output value of the cost function. Essentially, this function is "a function of the actual power of the environmental control equipment and the incentive price parameters".

[0059] In the game theory process, the occupant comfort utility function tends to drive the control scheme towards "high comfort, high power," while the energy cost function tends to drive the scheme towards "low energy consumption, low cost." The goal of solving the non-cooperative game is to find a device operation strategy that, under the hard constraint of an adjustable energy budget, makes it difficult for either party to gain a higher overall benefit (or a lower overall loss) by unilaterally changing their strategy. This strategy corresponds to the Nash equilibrium in the game model, which is a Pareto optimal compromise between occupant comfort and energy cost under the current budget and price conditions.

[0060] By completing step S3 above, the guest room local controller has successfully transformed the energy consumption control parameters from the upper layer into energy consumption constraint parameters and energy consumption cost coefficients in the local non-cooperative game model. Based on this, the guest comfort utility function and the guest room energy consumption cost function are clearly defined. Thus, the local decision model is tightly coupled with the hotel's overall optimization objective, laying a solid foundation for solving and executing specific equipment control commands within the framework of this game model in step S4.

[0061] S4: The guest room local controller, based on the adjusted non-cooperative game model, solves for and executes the optimal control command for the environmental control equipment in the guest room.

[0062] The predicted indoor environmental parameters determined by the operating settings of the environmental control equipment are within the comfort range determined based on the user preference data; the guest room local controller obtains the optimal operating settings of the environmental control equipment by solving the Nash equilibrium of the non-cooperative game model under the constraints, and generates corresponding equipment control commands.

[0063] The equipment control commands include at least one of the following parameters: the air conditioner's temperature setpoint, operating mode, fan speed setting, and fresh air valve opening.

[0064] Specifically, after adjusting the non-cooperative game model in step S3, the guest room local controller executes step S4. Its core task is to solve for the Nash equilibrium strategy of the non-cooperative game model under the hard constraints of ensuring the basic comfort of guests and without exceeding the upper-level energy consumption budget constraints, and to transform this mathematically optimal solution into specific control commands that can drive physical equipment, thus completing the closed loop from intelligent decision-making to environmental regulation execution.

[0065] Before starting the local optimization solution, the guest room local controller first determines the comfort range of key environmental parameters for this guest room based on user preference data.

[0066] In a preferred embodiment, for the temperature parameter, if the guest's preferred setting is 24°C, the local controller can set the allowable comfort range to be between 22°C and 26°C according to preset rules; similar allowable ranges can also be set for other parameters such as humidity. This allowable comfort range is considered a hard constraint in the subsequent solution process. Within any control cycle, the predicted indoor environmental parameters (calculated using the local thermal model and equipment model) corresponding to any candidate equipment operation strategy must fall within the aforementioned comfort allowable range; otherwise, the strategy will not be considered feasible in subsequent solutions. This approach ensures from the outset that the basic comfort experience of guests is not compromised regardless of adjustments to energy-saving strategies, avoiding extreme discomfort caused by control measures aimed at energy conservation.

[0067] In this embodiment, the guest room local controller cyclically executes the local decision-making and control process according to a fixed control cycle (e.g., every 15 minutes). Within each control cycle, the local controller sequentially completes the following steps: Information preparation and candidate solution generation; At the start of the current control cycle, the guest room local controller first completes the following information preparation: Read the remaining adjustable energy consumption budget for this guest room within the current demand response execution time window, or the corresponding time-of-use power constraint boundary; Obtain the latest measured indoor environmental parameters (such as temperature, humidity, carbon dioxide concentration, illuminance, etc.) and guest room occupancy information; Invoke the non-cooperative game model that has been adjusted in step S3, including the current expression and parameters of the guest comfort utility function and the room energy consumption cost function.

[0068] Based on the above information, the local controller generates several candidate control schemes for the current control cycle. Each candidate scheme corresponds to a set of operating setting parameters for the environmental control equipment, for example: The temperature setting of the air conditioner (a number of discrete points near the user's preferred value); Air conditioner operating modes (such as cooling, heating, dehumidification, and ventilation); Air conditioner fan speed setting (e.g., low, medium, high or automatic). Fresh air valve opening (e.g., several percentage settings); Optional lighting brightness or scene mode, curtain opening / closing status (as described in this embodiment).

[0069] Feasibility screening under dual constraints; For each candidate control scheme, the guest room local controller uses a local simplified thermal model and equipment power model to predict the changes in indoor environmental parameters and energy consumption that the scheme will cause during the current control cycle, and performs feasibility screening with dual constraints based on this prediction. Comfort Constraint Screening: Check whether the predicted indoor environmental parameters fall within the allowable comfort range determined based on user preference data. If any key parameter such as predicted temperature or humidity exceeds this range, the candidate solution is considered to not meet the hard comfort constraints and is eliminated.

[0070] Energy consumption budget constraint screening: Based on the remaining adjustable energy consumption budget (or time-sharing power constraint) of the current control cycle, calculate the energy consumption adjustment (or power level) brought about by implementing this scheme, and add the energy consumption already generated in the previous control cycle. Determine whether implementing this scheme will cause the cumulative energy consumption to exceed the energy consumption budget boundary allocated to this guest room by the upper-level collaborative model, or violate the upper and lower power limits constraints of this cycle. If it exceeds, the scheme is considered not to meet the energy consumption budget constraint and is also eliminated.

[0071] After screening under the above two types of constraints, the remaining candidate control schemes constitute the feasible strategy set for the current control cycle. Each strategy in this set is guaranteed to be within the comfort range and will not exceed the energy consumption budget allocated by the upper layer.

[0072] Solving Nash equilibrium in non-cooperative games; Within the set of feasible strategies (which simultaneously satisfies the comfort allowable range constraint and the remaining energy consumption budget constraint for this period), the guest room local controller solves for the corresponding Nash equilibrium strategy based on the non-cooperative game model adjusted in step S3.

[0073] From the perspective of the local controller, this process can be understood as finding an optimal compromise between the two objectives of "guest comfort utility" and "hotel energy cost" within a given energy budget. Specifically: For each feasible control scheme, the local controller calls the guest comfort utility function to calculate the deviation of the predicted indoor environmental parameters under that scheme from the user preference settings, and thus obtains the corresponding comfort utility value. At the same time, the guest room energy cost function is invoked to calculate the corresponding energy cost value based on the predicted power or energy consumption and incentive price parameters of the scheme in the current period.

[0074] In a preferred embodiment, the guest room local controller can employ iterative optimal response, weight adjustment, or other numerical game theory methods to perform game calculations on the set of feasible strategies. The basic idea is as follows: The "guest side" is viewed as a virtual participant seeking to maximize comfort utility, while the "energy consumption side" is viewed as a virtual participant seeking to minimize energy consumption costs. Within the set of feasible strategies, the preferences and selection processes of the two participants for each option are simulated, allowing each party to continuously revise its strategy in multiple rounds of "alternating responses". When a control scheme is found such that, without violating the allowable range of comfort and the energy consumption budget constraint, neither the guest side nor the energy consumption side can significantly improve their own benefits by deviating from the scheme unilaterally, it is considered that the corresponding operating setting parameters of the scheme have reached the Nash equilibrium of the local non-cooperative game model.

[0075] The set of equipment operating settings parameters corresponding to this Nash equilibrium strategy (such as the target temperature setpoint, fan speed, and fresh air valve opening for this cycle) are determined as the optimal operating settings parameters for the current control cycle.

[0076] In other embodiments, the local controller in the guest room can also approximate the above game problem as a problem of maximizing overall utility or minimizing overall cost. By setting appropriate weights to weight comfort utility and energy consumption cost, and using conventional optimization algorithms to solve the problem, this invention does not limit the scope of the invention.

[0077] Generation, issuance, and execution of equipment control commands; After determining the optimal operating settings for the current control cycle, the guest room local controller needs to transform these abstract parameters into specific control commands that the underlying physical devices can recognize and execute.

[0078] In a preferred embodiment: For air conditioning equipment, the local controller encodes parameters such as the optimal temperature setpoint, operating mode, and fan speed into control messages according to the communication protocol supported by the air conditioner, and sends them to the corresponding indoor unit or gateway via a bus such as RS-485, TCP / IP network, or manufacturer-specific bus. For fresh air equipment, the local controller converts the optimal fresh air valve opening into a corresponding analog output (such as a 0–10V voltage signal) or digital control command to drive the electric air valve or fresh air unit to adjust the fresh air volume. If this embodiment simultaneously controls other environmental control devices such as lighting and curtains, the local controller will also generate corresponding brightness settings, scene modes, or opening / closing control commands.

[0079] The aforementioned control commands are sent to each environmental control device in real time via the hardware interface module of the local controller. The devices adjust their operating status accordingly, thereby implementing the optimal operating settings obtained in step S4 at the physical level.

[0080] Furthermore, in one optional implementation, after issuing and executing optimal control commands, the guest room local controller can continuously monitor the actual operating status of the equipment and changes in indoor environmental parameters throughout the current control cycle and the entire demand response period, and compare these with previous model prediction results. The monitoring content may include, but is not limited to: Real-time or periodic reporting of the actual power, operating mode and set parameters of each environmental control device; The actual values ​​of indoor temperature, humidity, carbon dioxide concentration, illuminance, etc., collected by environmental sensors; Guests can manually intervene via wall panels, mobile devices, etc. (such as actively raising / lowering the temperature).

[0081] After the demand response event concludes, the local controller can package and organize the aforementioned operational data and interaction logs, and upload them to the central controller via the hotel's internal network. The central controller can then utilize this feedback data: Calibrate and update the user preference model to better align with guests' actual habits when building comfort limits and preference settings in the future; Update the historical energy consumption database and optimize the baseline load forecast and upper-level collaborative model parameters for subsequent demand response events; The parameters of the upper-level collaborative optimization algorithm and the local game model are iteratively adjusted to improve the stability and economy of future solutions.

[0082] Through the optional closed-loop mechanism of "execution-feedback-learning" described above, the system of the present invention can continuously accumulate experience in multiple demand response events, and realize the adaptive optimization and long-term evolution of the control strategy.

[0083] Example 2 is the second embodiment of the present invention, which differs from the previous embodiment in that: If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art or the current technical solution, can be embodied in the form of a software product. This current computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0084] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0085] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0086] Example 3 is an embodiment of the present invention, which provides a smart hotel room intelligent control system based on multi-source data, including a central controller and multiple local room controllers that are communicatively connected to the central controller; like Figure 3 As shown, the central controller 200 specifically includes: Signal receiving module 210 is used to receive demand response signals from the power grid-side dispatching platform or demand response platform; Data aggregation module 220 is used to receive and aggregate room operating status parameters from the local controller of each room; The collaborative optimization module 230 is used to construct an upper-level collaborative model with the goal of minimizing overall energy consumption cost based on the demand response signal and the operating status parameters, and to solve the model to generate energy consumption control parameters corresponding to each guest room. The parameter distribution module 240 is used to distribute the energy consumption control parameters for each guest room generated by the collaborative optimization module 230 to the corresponding guest room local controllers 300-1 to 300-N through the hotel's internal network.

[0087] like Figure 4 As shown, each guest room local controller 300 may functionally include: The parameter receiving module 310 is used to receive energy consumption control parameters from the central controller 200 and store them in the local configuration area. The local decision module 320 has a preset non-cooperative game model, which is used to adjust the energy consumption constraint parameters and energy consumption cost coefficient in the non-cooperative game model based on the energy consumption control parameters, and solve for the optimal operating setting parameters of the environmental control equipment in this guest room under the constraints. The control execution module 330 is used to convert the optimal operating setting parameters output by the local decision module 320 into specific equipment control commands, and to send and execute them through the communication interface with the air conditioning terminal, fresh air equipment and optional environmental control equipment such as lighting and curtains, so as to realize intelligent control of the environment of this guest room.

[0088] Example 4 is an embodiment of the present invention, which provides a smart hotel room intelligent control method based on multi-source data. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and comparative experiments.

[0089] This embodiment involved a field trial conducted in an operational smart business hotel. Located in a region with high summer temperatures, the hotel has 100 standard guest rooms. The central air conditioning and guest room terminals are all connected to the building management system, enabling room-by-room monitoring and control. The trial selected 100 guest rooms as observation subjects and conducted the test on a typical hot summer day. On that day, the outdoor maximum temperature was approximately 35°C. The power grid company initiated a demand response event between 2:00 PM and 4:00 PM, requiring the hotel to reduce its peak air conditioning electricity consumption by approximately 100 kW within this two-hour window, and offered a demand response incentive price of 1.50 yuan / kWh.

[0090] The central controller in this invention is deployed on an industrial server in the hotel's server room. It connects to the switches in the low-voltage rooms on each floor via the hotel's internal network and then communicates with the local controller in each guest room. Each guest room's local controller is a multi-protocol smart IoT gateway, connected to indoor environmental sensors (temperature, humidity, carbon dioxide concentration, illuminance), occupancy sensing devices (human presence sensors, smart door lock events), and environmental control devices (variable frequency air conditioning terminals, fresh air valves, and smart lighting in some rooms). Before the experiment began, the central controller continuously collected and aggregated multi-source operating status parameters according to step S1 of this invention, forming a complete status profile of each guest room before 14:00 that day. At the same time, it retrieved recent check-in records from the hotel's property management system and membership system, forming guests' preferred temperature ranges (such as "prefers cool," "standard," "prefers warm," etc.) and corresponding parameters.

[0091] To generate meaningful test results, this embodiment employs three different control strategies to conduct demand response events for the same hotel on multiple similar high-temperature days, and statistically compares key indicators: Baseline Scheme A: Traditional fixed setting control.

[0092] Without participating in demand response, all air conditioning units in guest rooms will maintain a constant temperature of 24°C, only adjusting their start / stop and frequency conversion based on standard room temperature feedback. This scheme primarily serves as a benchmark for comfort and energy consumption levels.

[0093] Compare with Option B: Traditional unified demand response control.

[0094] After the power grid issues a demand response signal, the central controller, following existing industry practices, uniformly issues control commands to all participating rooms, raising the air conditioning set temperature from 24°C to 26°C. The fresh air volume is not adjusted differently, and there is no consideration for whether the room is occupied or the temperature preferences of the guests. The peak load reduction is achieved simply by uniformly raising the set temperature.

[0095] Scheme C of this invention: a two-layer intelligent control method of collaborative optimization + local game theory.

[0096] After the power grid issues a demand response signal, the central controller, according to step S1 of this invention, obtains the demand response signal and the latest operating status parameters of each guest room. Following step S2, it constructs an upper-level collaborative model with the objective function of minimizing the sum of "energy consumption cost + comfort compensation cost." Using the target load adjustment amount and time window given by the power grid as conditions, it employs a distributed optimization algorithm to solve the problem, differentially allocating peak shaving tasks to each guest room and generating an adjustable energy consumption budget and corresponding incentive price parameters for each room. Subsequently, according to step S3, each guest room's local controller internalizes this budget and incentive price into energy consumption constraint parameters and energy consumption cost coefficients in the local non-cooperative game model. Combining the room's real-time environmental parameters and guest preference data, it constructs a guest comfort utility function and a guest room energy consumption cost function. Finally, according to step S4, under the dual constraints of a preset comfort allowable range (generally set to preferred temperature ±1.5℃ in this embodiment) and not exceeding the upper-level energy consumption budget, the local controller performs adjustments in each control cycle (15 cycles in this embodiment). (Minutes) Solve the Nash equilibrium of the non-cooperative game model to obtain the optimal air conditioning temperature setpoint, operating mode and fan speed level for that period, and convert them into control commands that the equipment can recognize and issue for execution.

[0097] During the aforementioned testing process, the central controller and local controllers continuously recorded the hotel's overall power load, temperature changes in each guest room, energy consumption adjustments, and guest service feedback information (including temperature-related complaints or dissatisfaction records) under each scheme. To enhance the generalizability of the results, this embodiment implemented Scheme C of the present invention under different occupancy rates (high occupancy rate approximately 85%, medium occupancy rate approximately 70%, and low occupancy rate approximately 50%), and finally conducted statistical and comparative analysis on the key indicators of the three control strategies. Please refer to Table 1 for specific data.

[0098] Table 1: Experimental Results Data Table

[0099] It should be noted that the "peak hour average indoor temperature" is the arithmetic mean of the measured room temperature of all guest rooms during the period from 14:00 to 16:00. "Percentage of time exceeding comfort limits" refers to the proportion of the cumulative time during which the measured indoor temperature exceeds its respective comfort range to the total duration of that period. "Average electricity savings per room" refers to the average electricity savings per room relative to the baseline plan A (not participating in demand response); "Overall hotel load reduction ratio" refers to the actual load reduction ratio for the hotel's air conditioning system. "Customer complaint rate" is the percentage of valid customer complaints directly related to room temperature comfort during the test day out of the total number of rooms involved. The "Comprehensive Revenue Per Room" is calculated by taking into account savings in electricity costs, demand response incentive revenue, and compensation costs and customer complaint risks that may result from a decrease in comfort.

[0100] As can be seen from the above field test data, the method of the present invention is significantly better than the traditional unified control scheme in terms of multi-dimensional performance indicators. At the same time, it maintains a comfortable experience close to the benchmark operation as much as possible while saving energy, which fully proves the rationality and inventiveness of the technical solution of the present invention.

[0101] Firstly, regarding guest comfort, the method of this invention effectively alleviates the typical contradiction of "energy saving versus experience" in traditional demand response strategies. Under the baseline scheme that does not participate in demand response, the average indoor temperature during peak hours is 24.1℃, and the percentage of time exceeding the comfort limit is 3.5%, which can be regarded as the comfort baseline of the hotel under normal operation on a high-temperature day. Traditional uniform demand response control raises the set temperature to 26℃ in a "one-size-fits-all" manner, resulting in a significant increase in the average room temperature during peak hours to 25.8℃, and a surge in the percentage of time exceeding the comfort limit to 18.7%. This indicates that the room temperature deviates significantly from the user's preferred range for a considerable period of time, and guests' subjective discomfort increases significantly. In comparison, the average indoor temperature of the method of this invention in high, medium, and low occupancy scenarios is 24.4℃, 24.2℃, and 24.0℃, respectively, with an overall average of only 24.2℃, which is very close to the benchmark solution's 24.1℃. The percentage of comfort boundary violations in the three scenarios is 5.2%, 4.1%, and 3.8%, respectively, with an average of 4.4%, which is only slightly higher than the benchmark level, but far lower than the 18.7% of the traditional unified control scheme. This indicates that under the dual mechanism of "hard constraints on the allowable comfort range + comfort compensation cost" adopted in this invention, the system prioritizes the protection of sensitive guest rooms when performing demand response tasks, and mainly focuses temperature adjustments on vacant rooms or rooms that are not sensitive to temperature, thereby achieving load reduction while basically maintaining a comfortable experience close to that of normal operation.

[0102] Secondly, regarding energy saving and peak shaving effects, the method of this invention achieves higher energy efficiency and a more reasonable load sharing structure while meeting the grid's peak shaving target. Traditional unified demand response control achieved an average energy saving of 2.1 kWh per room in this field test, corresponding to a reduction of approximately 14.3% in the overall hotel's air conditioning load. In actual tests with high, medium, and low occupancy rates, the method of this invention achieved average energy savings of 2.4 kWh, 2.7 kWh, and 3.0 kWh per room, respectively, with an overall average of 2.7 kWh, representing an improvement of approximately 28% compared to the traditional unified control scheme. The corresponding reductions in the overall hotel's air conditioning load were 15.2%, 15.9%, and 16.4%, respectively, with an overall average of 15.8%, significantly better than the traditional scheme's 14.3%. Combined with the grid's given peak shaving target of 100 kW, the method of this invention enables the actual load reduction for the entire hotel to approach or slightly exceed the target value in all three occupancy rate scenarios, with the reduction primarily borne by vacant rooms and rooms allowing for greater deviation. This directly demonstrates the advantage of the upper-level collaborative model in step S2 for differentiated budget allocation. By introducing comfort compensation costs and operational allowable range constraints, the question of "who should cut costs and how much should be cut" is transformed into a quantifiable optimization decision, thereby making the overall energy-saving effect better than traditional simple rules.

[0103] Furthermore, regarding operational risks and user satisfaction, the method of this invention effectively reduces the potential negative impacts of demand response. Test results show that the customer complaint rate under the baseline scheme is approximately 2.8%, reflecting the hotel's normal service level when not participating in demand response. When using traditional unified demand response control, the customer complaint rate rises to 9.6%, mainly concentrated in high-floor rooms, west-facing rooms, and for guests who prefer cooler or sensitive rooms, indicating that the unified heating strategy poses significant experience risks in practical applications. In contrast, the customer complaint rates of the method of this invention are 3.4%, 2.9%, and 2.5% in high, medium, and low occupancy scenarios, respectively, with an average of 2.6% across the three scenarios. This is not only far lower than the unified control strategy but also comparable to or even slightly improved upon the baseline scheme. This result demonstrates that by balancing guest preferences and energy costs through a lower-level non-cooperative game model and by imposing mandatory constraints on the allowable comfort range, the method of this invention can control user dissatisfaction and complaints at levels close to normal operations while ensuring the effectiveness of demand response, demonstrating a clear advantage from an operational risk control perspective.

[0104] Finally, in terms of overall economic benefits, the method of this invention also demonstrates significant advantages. Traditional unified demand response control yielded approximately RMB 11.2 per room in this test; the method of this invention achieved RMB 13.1, RMB 13.8, and RMB 14.4 per room in high, medium, and low occupancy scenarios, respectively, with an average of RMB 13.8 across the three scenarios, representing an improvement of approximately 23% compared to traditional unified control. This overall benefit considers both demand response incentive revenue and electricity cost savings, while deducting compensation costs and indirect losses that may arise from comfort deviations. Therefore, under the same grid incentive conditions, this invention, through a two-layer structure of "upper-level collaborative optimization + lower-level game theory decision-making," achieves multi-objective synergy between resource allocation and comfort assurance at the technical level, and achieves higher net revenue per room at the economic level.

[0105] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A smart hotel room intelligent control method based on multi-source data, applied to a system including a central controller and multiple local room controllers communicatively connected to the central controller, characterized in that, Includes the following steps: The central controller receives power grid demand response signals and acquires the operating status parameters of each guest room; Based on the demand response signal and the operating status parameters, the central controller constructs an upper-level collaborative model with the goal of minimizing overall energy consumption costs, solves and generates energy consumption control parameters for each guest room, and distributes the energy consumption control parameters to the local controllers of each guest room respectively. The guest room local controller receives the energy consumption control parameters for the guest room and adjusts its internal preset non-cooperative game model based on the energy consumption control parameters. The guest room local controller, based on the adjusted non-cooperative game model, solves for and executes the optimal control commands for the environmental control equipment in the guest room.

2. The intelligent control method for smart hotel rooms based on multi-source data as described in claim 1, characterized in that, The central controller receives power grid demand response signals including: The central controller receives demand response signals periodically or when a demand response event is triggered through a communication connection established with the grid-side dispatch platform or demand response platform; the demand response signal includes at least the target load adjustment amount, the demand response execution time window, and incentive price parameters.

3. The intelligent control method for smart hotel rooms based on multi-source data as described in claim 2, characterized in that, The operating status parameters include: Temperature, humidity, carbon dioxide concentration, and illuminance parameters were collected by environmental sensors in the guest rooms; Guest room occupancy status information obtained from guest room human body sensors and door lock systems; The operating mode, set parameters, and real-time power reported by the guest room environment control equipment; Historical energy consumption data and user preference data associated with guest rooms.

4. The intelligent control method for smart hotel rooms based on multi-source data as described in claim 3, characterized in that, The construction of the upper-level collaborative model with the goal of minimizing overall energy consumption cost includes: Based on the target load adjustment amount, demand response execution time window, and incentive price parameter in the demand response signal, an upper-level collaborative model is constructed with the objective function of minimizing the hotel's total electricity cost within the time window; wherein, the total electricity cost includes the energy consumption cost determined based on the incentive price parameter and the load adjustment amount of each guest room, as well as the compensation cost caused by the load adjustment leading to the indoor environment deviating from the baseline comfort level. The operating status parameters are used as constraints, which include at least the following: the sum of the load adjustment amounts of each guest room meets the target load adjustment amount, and the load adjustment amount of each guest room is within its corresponding operating allowable range.

5. The intelligent control method for smart hotel rooms based on multi-source data as described in claim 4, characterized in that, The energy consumption control parameters for each guest room are generated by solving the following: A distributed optimization algorithm is used to solve the upper-level collaborative model to obtain the adjustable energy consumption budget allocated to each guest room within the demand response execution time window; The adjustable energy consumption budget is combined with the incentive price parameter to generate energy consumption control parameters for each guest room, which are then sent to the corresponding guest room local controller via the hotel's internal communication network.

6. The intelligent control method for smart hotel rooms based on multi-source data as described in claim 5, characterized in that, The method of adjusting its internally preset non-cooperative game model based on the energy consumption control parameters includes: The adjustable energy consumption budget is used as the energy consumption constraint parameter in the non-cooperative game model, and the incentive price parameter is used as the energy consumption cost coefficient in the non-cooperative game model.

7. The intelligent control method for smart hotel rooms based on multi-source data as described in claim 6, characterized in that, The non-cooperative game model includes: Guest comfort utility function and room energy cost function; The guest comfort utility function is a function of the deviation between the actual indoor environmental parameters and the user preference settings, and the guest room energy cost function is a function of the actual power of the environmental control equipment and the incentive price parameter.

8. The intelligent control method for smart hotel rooms based on multi-source data as described in claim 7, characterized in that, The optimal control commands for the environmental control equipment in this guest room obtained by solving the problem include: The predicted indoor environmental parameters determined by the operating settings of the environmental control equipment are within the comfort range determined based on the user preference data; the guest room local controller obtains the optimal operating settings of the environmental control equipment by solving the Nash equilibrium of the non-cooperative game model under the constraints, and generates corresponding equipment control commands.

9. The intelligent control method for smart hotel rooms based on multi-source data as described in claim 8, characterized in that, The equipment control commands include at least one of the following parameters: air conditioner temperature setting, operating mode, fan speed setting, and fresh air valve opening.

10. A smart hotel room intelligent control system based on multi-source data, used to implement the smart hotel room intelligent control method based on multi-source data as described in any one of claims 1 to 9, characterized in that, include: A central controller and multiple guest room local controllers that are communicatively connected to the central controller; The central controller includes: Signal receiving module: configured to receive demand response signals from the power grid side; Data aggregation module: configured to acquire the operating status parameters of each guest room; Collaborative optimization module: configured to construct an upper-level collaborative model with the goal of minimizing overall energy consumption cost based on the demand response signal and the operating status parameters, and solve the model to generate energy consumption control parameters for each guest room; Parameter distribution module: configured to distribute the energy consumption control parameters for each guest room generated by the solution to the corresponding local controller of the guest room; The guest room local controller includes: Parameter receiving module: configured to receive energy consumption control parameters from the central controller; Local decision-making module: It has a preset non-cooperative game model and is configured to adjust the non-cooperative game model based on the energy consumption control parameters, and solve for the optimal control command of the environmental control equipment in this guest room; Control execution module: configured to execute the optimal control command to control the environmental control equipment.

Images