Intelligent building energy consumption regulation method and system based on multi-source data fusion

Abstract

The application relates to the technical field of building energy consumption management, and relates to an intelligent building energy consumption regulation and control method and system based on multi-source data fusion, which comprises the following steps: confirming an initial building, confirming an external environment detection mechanism, evaluating the conditions of multiple internal rooms, obtaining multiple optimal condition groups, modeling behaviors of multiple historical parameter groups, obtaining multiple personnel moment groups and multiple energy consumption regulation and control models, performing particle swarm simulation on the multiple energy consumption regulation and control models based on the multiple optimal condition groups, multiple prediction disturbance indexes, current light intensity, current wind speed and current external temperature, obtaining multiple optimal machine parameter groups, setting parameters of the multiple internal rooms in the initial building by using the multiple optimal machine parameter groups, obtaining a target building, and completing energy consumption regulation and control of the building. The application can improve the comfort of personnel in the building and reduce building energy consumption.

Description

Technical Field

[0001] This invention relates to the field of building energy management technology, and in particular to an intelligent building energy consumption control method and system based on multi-source data fusion. Background Technology

[0002] With the development of smart cities and green buildings, building energy consumption control technology is gradually evolving from traditional manual adjustment and timed control methods towards automation and intelligence. In multi-functional buildings such as office buildings and industrial plants, how to achieve precise energy saving while ensuring indoor comfort has become an important research topic in intelligent building energy consumption regulation.

[0003] In existing smart building systems, multiple sensors are typically deployed to monitor the building’s internal environment and external climate parameters. Some systems also introduce rule-based control logic or preliminary model predictive control methods to coordinate and control high-energy-consuming equipment such as air conditioning and lighting.

[0004] While existing technologies can regulate building energy consumption, most rely on static rules or empirical models, failing to fully integrate the dynamic correlations of multi-source data across time and space. They lack the ability to predict occupant usage patterns and intelligent adaptive mechanisms to complex disturbances. In practical applications, this often results in insufficient control precision, delayed response, or an inability to balance comfort and energy efficiency. Therefore, there is an urgent need for a dynamic building energy consumption adjustment method that integrates multi-source sensing information and intelligent optimization algorithms to improve occupant comfort and reduce building energy consumption. Summary of the Invention

[0005] This invention provides a smart building energy consumption control method based on multi-source data fusion and a computer-readable storage medium, the main purpose of which is to improve the comfort of people in the building and reduce building energy consumption.

[0006] To achieve the above objectives, the present invention provides a smart building energy consumption control method based on multi-source data fusion, comprising:

[0007] The initial building has been identified, which includes: multiple internal rooms, and each internal room includes: a multi-source sensing mechanism, a variable temperature air conditioner, and a multi-level lighting device. The multi-source sensing mechanism includes: a people recognition device, an access detector, an infrared camera, a brightness sensor, a humidity sensor, and a temperature sensor.

[0008] The external environment detection mechanisms have been identified, including: wind speed sensor, light sensor and external temperature sensor;

[0009] The suitability conditions of multiple internal rooms were evaluated to obtain multiple optimal condition groups. Each optimal condition group corresponds to an internal room and includes: optimal temperature, optimal humidity, and optimal brightness.

[0010] Behavioral modeling is performed on multiple pre-constructed sets of historical parameters to obtain multiple sets of personnel time and multiple energy consumption control models. The sets of historical parameters, personnel time, and energy consumption control models are all one-to-one correspondences with the internal rooms.

[0011] When a pre-built energy consumption control instruction is received, the current time is determined, and multiple predicted disturbance indices are determined in multiple personnel time groups based on the current time.

[0012] The wind speed sensor, light sensor and external temperature sensor in the external environment detection mechanism are used to obtain the current wind speed, current light intensity and current external temperature respectively;

[0013] Based on multiple sets of optimal conditions, multiple predicted disturbance indices, current light intensity, current wind speed, and current external temperature, particle swarm simulation was performed on multiple energy consumption control models to obtain multiple sets of optimal machine parameters.

[0014] The system uses multiple optimal machine parameter sets to set parameters for multiple internal rooms in the initial building to obtain the target building. Starting from the time when the target building is obtained, the system records the time in real time to obtain the stage time. The stage time is continued until the stage time reaches the preset stage threshold. The target building is then used as the initial building, and the system returns to the step of confirming the current time. This process continues until a pre-constructed control end command is received, thus completing the energy consumption control of the building.

[0015] Optionally, the assessment of suitability conditions for multiple internal rooms yields multiple sets of optimal conditions, including:

[0016] Perform the following operation on each of the multiple internal rooms:

[0017] Obtain the historical assessment dataset of the internal rooms. The historical assessment dataset includes multiple historical assessment data points, including assessment temperature, assessment brightness, assessment humidity, and assessment scores.

[0018] Sort multiple historical assessment data in the historical assessment dataset according to the assessment scores from high to low to obtain the historical assessment data sequence;

[0019] Confirm the number of assessments for multiple historical assessment data points in the historical assessment dataset;

[0020] The target number is determined based on the preset optimal ratio and the number of evaluations, where the target number is the product of the optimal ratio and the number of evaluations.

[0021] Based on the target quantity, multiple optimal evaluation data are extracted from the historical evaluation data sequence;

[0022] Multiple evaluation temperatures, multiple evaluation brightness, and multiple evaluation humidity were extracted from multiple optimal evaluation data;

[0023] The optimal temperature is calculated based on multiple measured temperatures, where the optimal temperature is the average of the multiple measured temperatures.

[0024] The optimal brightness is calculated based on multiple brightness measurements, and the optimal humidity is calculated based on multiple humidity measurements.

[0025] The optimal conditions are obtained by combining the optimal temperature, optimal brightness, and optimal humidity.

[0026] By summarizing the optimal condition groups, multiple optimal condition groups are obtained.

[0027] Optionally, before performing behavioral modeling on multiple pre-constructed sets of historical parameters to obtain multiple sets of personnel time intervals and multiple energy consumption control models, the method further includes:

[0028] Perform the following operation on each of the multiple internal rooms:

[0029] The external wind speed, external light intensity, and external temperature are obtained by using the wind speed sensor, light sensor, and external temperature sensor in the external environment detection mechanism, respectively.

[0030] The brightness, humidity and temperature of the interior room are collected by the brightness sensor, humidity sensor and temperature sensor in the multi-source sensing mechanism, respectively.

[0031] Confirm the set temperature and fan speed of the variable temperature air conditioner in the room, and confirm the set brightness of the multi-level lighting device;

[0032] Use a people detector to identify the current number of people in the rooms;

[0033] The number of people entering and exiting the internal rooms is obtained based on the preset monitoring time interval and the access detectors.

[0034] Based on a preset shooting time, shooting time interval, and infrared camera, multiple infrared images of the interior room are acquired. The infrared images include multiple infrared pixels, and each infrared pixel includes an infrared temperature value.

[0035] For each of the multiple infrared images, perform the following operation:

[0036] Calculate the temperature variance based on multiple infrared pixels in the infrared image;

[0037] Summarize the temperature variances to obtain multiple temperature variances. Calculate the average temperature variance and the advanced temperature standard deviation based on the multiple temperature variances. The average temperature variance is the average of the multiple temperature variances, and the advanced temperature standard deviation is the standard deviation of the multiple temperature variances.

[0038] Infrared volatility is calculated based on the mean temperature variance and the advanced temperature standard deviation, where infrared volatility is the product of the mean temperature variance and the advanced temperature standard deviation.

[0039] The personnel disturbance index is calculated based on infrared volatility, the number of people entering and exiting, and the current number of people. The historical moment when the personnel disturbance index is calculated is recorded.

[0040] By summarizing historical data such as external wind speed, external light intensity, external temperature, set temperature, set wind speed, set brightness, room brightness, room humidity, room temperature, and occupant disturbance index, a historical parameter set is obtained.

[0041] The historical parameter group is integrated into a historical data packet and stored in a pre-built waiting memory to obtain the historical memory. The time of obtaining the historical memory is used as the starting point and the time is recorded in real time to obtain the recording time. When the recording time reaches a preset recording time threshold, the historical memory is used as a waiting memory, and the process of performing the following operations on each of the multiple internal rooms is returned until a pre-built collection end instruction is received, and multiple historical data packets are extracted from the historical memory.

[0042] Historical parameter sets are obtained by using multiple historical data packets, and these sets are then aggregated to obtain multiple historical parameter sets, each corresponding one-to-one with an internal room.

[0043] Optionally, the formula for calculating the personnel disturbance index is as follows:

[0044]

[0045] Where, τ P σ represents the personnel disturbance index. T Let N be the infrared fluctuation rate, N0 be the current number of people, and N be the number of people in the current population. x Let ln be the number of people entering and exiting, ln be the natural logarithm, and e be the natural constant.

[0046] Optionally, the step of performing behavioral modeling on multiple pre-constructed sets of historical parameters to obtain multiple sets of personnel time periods and multiple energy consumption control models includes:

[0047] Perform the following operation for each of the multiple sets of historical parameter sets:

[0048] Construct the overall condition matrix and overall result matrix using historical parameter sets;

[0049] The pre-constructed deep learning model is trained using the conditional total matrix and the result total matrix to obtain the energy consumption regulation model;

[0050] Perform the following operation on each historical parameter group in the historical parameter group set:

[0051] By combining the historical time and personnel disturbance index in the historical parameter group, a personnel time group is obtained;

[0052] Summarize the personnel time groups to obtain a personnel time group set, which includes multiple personnel time groups;

[0053] By summarizing the personnel time sets and energy consumption control models, multiple personnel time sets and multiple energy consumption control models are obtained.

[0054] Optionally, the step of constructing the conditional total matrix and the result total matrix using historical parameter sets includes:

[0055] Multiple external wind speeds, multiple external light intensities, multiple external temperatures, multiple set temperatures, multiple set wind speeds, multiple set brightness, multiple room brightness, multiple room humidity, multiple room temperatures, and multiple occupant disturbance indices were extracted from the historical parameter set.

[0056] Perform the following operation on each historical parameter group in the historical parameter group set:

[0057] The normalized external wind speed is obtained by normalizing the external wind speed in the historical parameter set based on multiple external wind speeds.

[0058] The following parameters are used to obtain the following information: normalized external light intensity, normalized external temperature, normalized set temperature, normalized set wind speed, normalized set wind speed, normalized set brightness, normalized room brightness, normalized room humidity, normalized room humidity, normalized room temperature, and normalized disturbance index.

[0059] Construct an initial condition matrix based on normalized external wind speed, normalized external light intensity, normalized external temperature, normalized design temperature, normalized design wind speed, normalized design brightness, and normalized disturbance index;

[0060] An initial matrix of results is constructed based on the brightness, humidity, and temperature of the unified room.

[0061] The initial condition matrix and the initial result matrix are combined to obtain the control training group;

[0062] By summarizing the control training groups, multiple control training groups were obtained;

[0063] Construct a conditional total matrix based on multiple control training groups;

[0064] A total result matrix was constructed based on multiple control training groups.

[0065] Optionally, the step of identifying multiple predicted disturbance indices based on the current time across multiple personnel time groups includes:

[0066] The current time range is determined based on the current moment and the preset reference duration;

[0067] Perform the following operation on each of the multiple personnel time groups:

[0068] Based on the current time range, multiple target personnel time groups were identified in the personnel time group set, where the historical times in the target personnel time groups are located within the current time range;

[0069] The predicted disturbance index is calculated based on multiple target personnel time groups, where the predicted disturbance index is the average of the personnel disturbance indices in multiple target personnel time groups;

[0070] By summing the predicted disturbance indices, multiple predicted disturbance indices are obtained, and each predicted disturbance index corresponds one-to-one with a set of personnel time periods.

[0071] Optionally, the step of performing particle swarm optimization on multiple energy consumption control models based on multiple sets of optimal conditions, multiple predicted perturbation indices, current light intensity, current wind speed, and current external temperature to obtain multiple sets of optimal machine parameters, including:

[0072] Confirm the fan speed and temperature selection ranges for the variable temperature air conditioner, and confirm the brightness selection ranges for the multi-level lighting device;

[0073] The first speed selection range is determined based on the wind speed selection range;

[0074] The second speed selection range is determined based on the temperature selection range, and the third speed selection range is determined based on the brightness selection range.

[0075] For each of the multiple energy consumption control models, perform the following operation:

[0076] The predicted disturbance index corresponding to the energy consumption control model among multiple predicted disturbance indices is denoted as the target disturbance index, and the optimal condition group corresponding to the energy consumption control model among multiple optimal condition groups is denoted as the target optimal condition group.

[0077] Based on multiple room brightness, multiple room humidity, multiple room temperature, and the optimal temperature, optimal humidity, and optimal brightness in the target optimal condition group, obtain the normalized target brightness, normalized target humidity, and normalized target temperature.

[0078] Obtain a particle swarm, wherein the particle swarm includes: multiple particles;

[0079] Perform the following operation on each particle in the particle swarm:

[0080] The wind speed selection range is randomly selected to obtain the position of the wind speed particles;

[0081] The positions of temperature particles, brightness particles, first particle velocities, second particle velocities, and third particle velocities are obtained based on the temperature selection range, brightness selection range, first velocity selection range, second velocity selection range, and third velocity selection range, respectively.

[0082] The three-dimensional position of the particles was determined based on the positions of wind speed particles, temperature particles, and brightness particles.

[0083] The three-dimensional velocity of the particle is determined based on the velocities of the first, second, and third particles.

[0084] The particles are labeled by setting their three-dimensional position and three-dimensional velocity.

[0085] By summing the labeled particles, we obtain a swarm of labeled particles;

[0086] Based on the current light intensity, current wind speed, current external temperature, energy consumption control model, normalized target brightness, normalized target humidity, normalized target temperature and target disturbance index, the labeled particle swarm is iterated to obtain the optimal set of machine parameters.

[0087] By summarizing the optimal machine parameter sets, multiple optimal machine parameter sets are obtained, and each optimal machine parameter set corresponds one-to-one with an energy consumption control model.

[0088] Optionally, the step of performing particle swarm iteration on the marked particle swarm based on the current light intensity, current wind speed, current external temperature, energy consumption control model, normalized target brightness, normalized target humidity, normalized target temperature, and target perturbation index to obtain the optimal machine parameter set includes:

[0089] Multiple particle memories were identified, and each particle memory corresponds one-to-one with a labeled particle in the labeled particle swarm.

[0090] For each labeled particle in the labeled particle swarm, perform the following operation:

[0091] The initial particle matrix is ​​obtained based on the current light intensity, current wind speed, current external temperature, target disturbance index, wind speed particle position, temperature particle position, brightness particle position in the three-dimensional position of the marked particle, multiple external wind speeds, multiple external light intensities, multiple external temperatures, multiple set temperatures, multiple set wind speeds, multiple set brightness, and multiple personnel disturbance indices.

[0092] The initial particle matrix is ​​input into the energy consumption control model to obtain the particle result matrix, which includes: predicted room brightness, predicted room humidity and predicted room temperature.

[0093] Particle fitness is calculated based on predicted room brightness, predicted room humidity, predicted room temperature, normalized target brightness, normalized target humidity, and normalized target temperature, using the following formula:

[0094]

[0095] Where, γ z For particle fitness, N ml N ms N mw R L R S and R W These are the normalized target brightness, normalized target humidity, normalized target temperature, predicted room brightness, predicted room humidity, and predicted room temperature, respectively, where | represents taking the absolute value;

[0096] The particle fitness and the three-dimensional position of the labeled particle are integrated into a particle data package, and the particle data package is stored in the particle memory corresponding to the particle data package to obtain the update memory.

[0097] The target data packet is identified based on the updated memory, where the target data packet is the particle data packet with the lowest particle fitness among all particle data packets in the updated memory;

[0098] The target data packets and update memories are summarized separately to obtain multiple target data packets and multiple update memories, where each target data packet and update memory corresponds one-to-one with a marked particle.

[0099] The optimal data packet is determined based on multiple target data packets. The optimal data packet is the target data packet with the smallest particle fitness among the multiple target data packets.

[0100] The global optimal position is determined based on the best data packet, where the global optimal position is the three-dimensional position of the particle in the best data packet;

[0101] For each of the multiple target data packets, perform the following operation:

[0102] The labeled particles corresponding to the target data packet are updated based on the three-dimensional position and the global optimal position of the particles in the target data packet, thus obtaining the updated particles;

[0103] Summarize the updated particles to obtain the updated particle swarm;

[0104] The global fitness is obtained based on the wind speed particle position, temperature particle position, brightness particle position, normalized target brightness, normalized target humidity and normalized target temperature in the global optimal position, and the global fitness is compared with the preset fitness threshold.

[0105] If the global fitness is greater than the fitness threshold, the updated particle swarm is used as the labeled particle swarm, the multiple update memories are used as multiple particle memories, and the step of performing the following operation on each labeled particle in the labeled particle swarm is returned until the global fitness is less than or equal to the fitness threshold.

[0106] If the global fitness is less than or equal to the fitness threshold, then the optimal wind speed, optimal temperature and optimal brightness are determined based on the global optimal position, where the optimal wind speed is the wind speed particle position in the global optimal position, the optimal temperature is the temperature particle position in the global optimal position, and the optimal brightness is the brightness particle position in the global optimal position.

[0107] Confirm the multiple settable fan speeds and multiple settable temperatures of the variable temperature air conditioner, and confirm the multiple settable brightness levels of the multi-level lighting device;

[0108] Based on the optimal wind speed, the nearest optimal wind speed is identified among multiple settable wind speeds; based on the optimal temperature, the nearest optimal temperature is identified among multiple settable temperature ranges; based on the optimal brightness, the nearest optimal brightness is identified among multiple settable brightness ranges.

[0109] The optimal set of machine parameters is obtained by summarizing the nearest optimal wind speed, nearest optimal temperature, and nearest optimal brightness.

[0110] To achieve the above objectives, the present invention also provides an intelligent building energy consumption control system based on multi-source data fusion, comprising:

[0111] The initial environment confirmation module is used to confirm the initial building, which includes multiple internal rooms, and each internal room includes a multi-source sensing mechanism, a variable temperature air conditioner, and a multi-level lighting device. The multi-source sensing mechanism includes a people recognition device, an access detector, an infrared camera, a brightness sensor, a humidity sensor, and a temperature sensor. The module also confirms the external environment detection mechanism, which includes a wind speed sensor, a light sensor, and an external temperature sensor.

[0112] The historical parameter modeling module is used to evaluate the suitability of multiple internal rooms and obtain multiple optimal condition groups. Each optimal condition group corresponds one-to-one with an internal room and includes optimal temperature, optimal humidity, and optimal brightness. The module performs behavioral modeling on the pre-constructed sets of historical parameter groups to obtain multiple sets of personnel time and multiple energy consumption control models. Each set of historical parameter groups, personnel time groups, and energy consumption control models corresponds one-to-one with an internal room.

[0113] The building parameter optimization module is used to determine the current time when a pre-built energy consumption control command is received. Based on the current time, it determines multiple predicted disturbance indices from multiple personnel time groups. It uses wind speed sensors, light sensors, and external temperature sensors in the external environment detection mechanism to obtain the current wind speed, current light intensity, and current external temperature, respectively. Based on multiple optimal condition groups, multiple predicted disturbance indices, current light intensity, current wind speed, and current external temperature, it performs particle swarm simulation on multiple energy consumption control models to obtain multiple optimal machine parameter groups.

[0114] The energy consumption control loop module is used to set parameters for multiple internal rooms in the initial building using multiple optimal machine parameter sets to obtain the target building. Starting from the time of obtaining the target building, the time is recorded in real time to obtain the stage time. Until the stage time reaches the preset stage threshold, the target building is used as the initial building, and the process returns to the step of confirming the current time. The process continues until a pre-constructed control end command is received, thus completing the energy consumption control of the building.

[0115] To address the above problems, the present invention also provides an electronic device, the electronic device comprising:

[0116] Memory, storing at least one instruction;

[0117] The processor executes the instructions stored in the memory to implement the above-described intelligent building energy consumption control method based on multi-source data fusion.

[0118] To address the aforementioned issues, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the aforementioned intelligent building energy consumption control method based on multi-source data fusion.

[0119] To address the problems described in the background art, this invention identifies an initial building, comprising multiple internal rooms, each including a multi-source sensing mechanism, a variable-temperature air conditioner, and multi-level lighting. The multi-source sensing mechanism includes a people detector, an access detector, an infrared camera, a brightness sensor, a humidity sensor, and a temperature sensor. This invention, by identifying the initial building, clarifies the deployment of various sensing devices within the multiple internal rooms, providing data support and a spatial basis for the implementation of subsequent energy consumption control strategies. Furthermore, it identifies an external environment detection mechanism, including a wind speed sensor, a light sensor, and an external temperature sensor. This invention, by identifying the external environment detection mechanism, thereby achieves… Dynamic perception of the building's external environment facilitates subsequent particle swarm simulation based on current wind speed, light intensity, and external temperature. This allows for the assessment of suitable conditions in multiple internal rooms, resulting in multiple optimal condition sets. Each optimal condition set corresponds one-to-one with an internal room and includes optimal temperature, optimal humidity, and optimal brightness. Therefore, this embodiment of the invention, by assessing suitable conditions in multiple internal rooms, clarifies the optimal environmental control target among them, facilitating subsequent particle swarm simulation with specific objectives based on the optimal condition sets. This improves the comfort of occupants within the building. Furthermore, behavioral modeling is performed on pre-constructed sets of historical parameters, resulting in multiple sets of occupant timeframes and multiple energy consumption control models. These historical parameter sets and occupant timeframes are used to model behavior, leading to multiple sets of occupant timeframes and multiple energy consumption control models. Each personnel time set and energy consumption control model corresponds one-to-one with an internal room. This invention demonstrates that by modeling the behavior of multiple historical parameter sets, multiple personnel time sets are obtained, reflecting the activity patterns of personnel within the initial building at different time periods. This leads to the construction of multiple energy consumption control models, which predict the environmental state of internal rooms under specific external environments, set temperatures, set wind speeds, and set brightness levels. This facilitates subsequent particle swarm simulation based on the energy consumption control models. When a pre-constructed energy consumption control command is received, the current time is determined. Based on this current time, multiple predicted disturbance indices are identified from the multiple personnel time sets. Wind speed sensors, light sensors, and external temperature sensors in the external environment detection mechanism are used to obtain... The current wind speed, current light intensity, and current external temperature demonstrate that this embodiment of the invention provides real-time external parameters for subsequent energy consumption control by confirming multiple predicted disturbance indices and current external environmental data. This improves the environmental suitability and accuracy of energy consumption control. Based on multiple optimal condition sets, multiple predicted disturbance indices, current light intensity, current wind speed, and current external temperature, particle swarm simulation is performed on multiple energy consumption control models to obtain multiple optimal machine parameter sets. This embodiment of the invention inputs multi-source data into the particle swarm algorithm for simulation calculation. By continuously trying multiple different temperature, wind speed, and brightness settings and substituting them into the energy consumption control model to calculate particle fitness, the optimal machine parameter set that best matches the optimal condition set is globally searched.This invention utilizes multiple optimal machine parameter sets to set parameters for multiple internal rooms in an initial building, resulting in a target building. Starting from the time the target building is obtained, the time is recorded in real-time to obtain a stage time. This stage time is repeated until a preset stage threshold is reached. The target building is then used as the initial building, and the process returns to the step of confirming the current time. This continues until a pre-constructed control end command is received, completing the energy consumption control of the building. Thus, this embodiment of the invention sequentially sets the variable-temperature air conditioning and multi-level lighting devices in each internal room of the initial building using multiple optimal machine parameter sets, thereby improving the comfort of occupants. Through a cyclical step of stage time construction, new optimal machine parameter sets are regenerated stage by stage based on the current environmental state, achieving staged adaptive optimization control. This effectively avoids energy efficiency losses caused by redundant energy output under mismatched environmental conditions, thereby reducing building energy consumption. Therefore, this invention can improve the comfort of occupants and reduce building energy consumption. Attached Figure Description

[0120] Figure 1 This is a flowchart illustrating an embodiment of the intelligent building energy consumption control method based on multi-source data fusion provided by the present invention.

[0121] Figure 2 A functional block diagram of an intelligent building energy consumption control system based on multi-source data fusion provided in an embodiment of the present invention;

[0122] Figure 3 This is a schematic diagram of the structure of an electronic device for implementing the intelligent building energy consumption control method based on multi-source data fusion, according to an embodiment of the present invention.

[0123] Explanation of reference numerals in the attached figures:

[0124] 10. Electronic device; 11. Processor; 12. Memory; 13. Bus.

[0125] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0126] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0127] This application provides a method for intelligent building energy consumption control based on multi-source data fusion. The executing entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0128] Reference Figure 1 The diagram shown is a flowchart illustrating a smart building energy consumption control method based on multi-source data fusion according to an embodiment of the present invention. In this embodiment, the smart building energy consumption control method based on multi-source data fusion includes:

[0129] S1. Identify the initial building, which includes: multiple internal rooms, and each internal room includes: a multi-source sensing mechanism, a variable temperature air conditioner, and a multi-level lighting device. The multi-source sensing mechanism includes: a people recognition device, an access detector, an infrared camera, a brightness sensor, a humidity sensor, and a temperature sensor.

[0130] It should be explained that the initial building refers to the building to be energy-controlled in this embodiment of the invention, such as an office building or industrial plant. An internal room refers to an independent spatial unit within the initial building, divided according to function, purpose, or area. For example, if the initial building is an office building, the internal rooms are conference rooms, offices, or rest areas; if the initial building is an industrial plant, the internal rooms are a production workshop. A variable-temperature air conditioner refers to an air conditioner installed in an internal room that can set its operating wind speed and temperature levels. For example, the wind speed levels include a first wind speed, a second wind speed, and a third wind speed, wherein the first wind speed is less than the second wind speed, and the second wind speed is less than the third wind speed. The specific wind speeds corresponding to the first, second, and third wind speeds are related to the model of the variable-temperature air conditioner. The temperature levels include 16℃, 17℃, ..., 30℃. A multi-level lighting device refers to a lighting device installed in an internal room that has multi-level brightness adjustment capabilities, allowing switching between different brightness levels. The multi-source sensing mechanism is a device that integrates a people detector, an entry / exit detector, an infrared camera, a brightness sensor, a humidity sensor, and a temperature sensor within the interior room. The people detector is a sensing device used to detect the total number of people currently present in the interior room. It is installed on the ceiling of the interior room; optionally, the Xingzong IoT VS121-P spatial people sensor is used as the people detector. The entry / exit detector is a sensor based on the principle of infrared beam transmission. Its transmitter and receiver are installed on opposite sides of the entrance to the interior room. When an object or person passes through the infrared rays emitted by the transmitter, it blocks the reception of the infrared signal by the receiver, thus generating a trigger signal for identifying personnel entry / exit behavior or counting the number of passages. Optionally, the Xingzong IoT VS360 entrance / exit people counting sensor is used as the entry / exit detector. The infrared camera is an infrared video camera installed in the top corner of the interior room to capture infrared images of the people and the interior environment from a top-down angle. A brightness sensor is used to monitor the light intensity in an interior room; a humidity sensor is used to monitor the relative humidity of the air in an interior room; and a temperature sensor is used to monitor the air temperature in an interior room.

[0131] S2. Identify the external environment detection mechanisms, which include: wind speed sensor, light sensor and external temperature sensor.

[0132] It should be explained that the external environment detection mechanism is a device integrating a wind speed sensor, a light sensor, and an external temperature sensor. These sensors are all installed at preset initial positions on the exterior surface of the initial building. The specific initial positions are manually set by staff at the energy consumption control center. For details on the application of the energy consumption control center, please refer to the following embodiments. The wind speed sensor is used to monitor the wind speed at the initial position outside the initial building; the light sensor is used to monitor the intensity of natural light at the initial position outside the initial building; and the external temperature sensor is used to monitor the air temperature at the initial position outside the initial building.

[0133] S3. Conduct a suitability assessment on multiple internal rooms to obtain multiple optimal condition groups. Each optimal condition group corresponds to one internal room and includes: optimal temperature, optimal humidity, and optimal brightness.

[0134] In detail, the assessment of suitability conditions for multiple internal rooms yields multiple sets of optimal conditions, including:

[0135] Perform the following operation on each of the multiple internal rooms:

[0136] Obtain the historical assessment dataset of the internal rooms. The historical assessment dataset includes multiple historical assessment data points, including assessment temperature, assessment brightness, assessment humidity, and assessment scores.

[0137] Sort multiple historical assessment data in the historical assessment dataset according to the assessment scores from high to low to obtain the historical assessment data sequence;

[0138] Confirm the number of assessments for multiple historical assessment data points in the historical assessment dataset;

[0139] The target number is determined based on the preset optimal ratio and the number of evaluations, where the target number is the product of the optimal ratio and the number of evaluations.

[0140] Based on the target quantity, multiple optimal evaluation data are extracted from the historical evaluation data sequence;

[0141] Multiple evaluation temperatures, multiple evaluation brightness, and multiple evaluation humidity were extracted from multiple optimal evaluation data;

[0142] The optimal temperature is calculated based on multiple measured temperatures, where the optimal temperature is the average of the multiple measured temperatures.

[0143] The optimal brightness is calculated based on multiple brightness measurements, and the optimal humidity is calculated based on multiple humidity measurements.

[0144] The optimal conditions are obtained by combining the optimal temperature, optimal brightness, and optimal humidity.

[0145] By summarizing the optimal condition groups, multiple optimal condition groups are obtained.

[0146] It should be explained that historical evaluation data is obtained by evaluating people in the internal rooms. For example, when a person is in a certain internal room, they are asked to rate the comfort level of the environment based on their own feelings (1-10 points, with higher scores indicating higher comfort). The rating score is the result of the rating. After the person completes the rating, the brightness, humidity, and temperature of the room are obtained using the brightness sensor, humidity sensor, and temperature sensor in the multi-source sensing mechanism, respectively, as the evaluation brightness, humidity sensor, and temperature sensor. Finally, the evaluation temperature, brightness, humidity, and rating score are summarized to obtain the historical evaluation data. The historical evaluation dataset is the collection of all historical evaluation data obtained from multiple different evaluations of the internal room over the years.

[0147] Understandably, the number of assessments refers to the number of historical assessment data points from multiple historical assessment datasets. For example, if the optimal ratio is 10% and the number of assessments is 100, then the target number is 10. Therefore, the top 10 historical assessment data points are extracted from the historical assessment data sequence as 10 optimal assessment data points. Since each optimal assessment data point includes assessment temperature, assessment brightness, and assessment humidity, multiple assessment temperatures, multiple assessment brightness levels, and multiple assessment humidity levels can be extracted from multiple optimal assessment data points. The optimal ratio is manually set by the staff of the energy consumption control center; preferably, the optimal ratio is 10%.

[0148] It should be understood that the methods for calculating the optimal brightness based on multiple brightness measurements and the methods for calculating the optimal humidity based on multiple humidity measurements are the same as the methods for calculating the optimal temperature based on multiple temperature measurements, and will not be described again here.

[0149] For example, if the optimal temperature is 25°C, the optimal brightness is 500 Lux, and the optimal humidity is 30%, then the optimal condition group is {25°C, 500 Lux, 30%}.

[0150] S4. Perform behavioral modeling on multiple pre-constructed sets of historical parameters to obtain multiple sets of personnel time and multiple energy consumption control models. The sets of historical parameters, personnel time, and energy consumption control models correspond one-to-one with the internal rooms.

[0151] Specifically, before performing behavioral modeling on multiple pre-constructed sets of historical parameters to obtain multiple sets of personnel time intervals and multiple energy consumption control models, the process also includes:

[0152] Perform the following operation on each of the multiple internal rooms:

[0153] The external wind speed, external light intensity, and external temperature are obtained by using the wind speed sensor, light sensor, and external temperature sensor in the external environment detection mechanism, respectively.

[0154] The brightness, humidity and temperature of the interior room are collected by the brightness sensor, humidity sensor and temperature sensor in the multi-source sensing mechanism, respectively.

[0155] Confirm the set temperature and fan speed of the variable temperature air conditioner in the room, and confirm the set brightness of the multi-level lighting device;

[0156] Use a people detector to identify the current number of people in the rooms;

[0157] The number of people entering and exiting the internal rooms is obtained based on the preset monitoring time interval and the access detectors.

[0158] Based on a preset shooting time, shooting time interval, and infrared camera, multiple infrared images of the interior room are acquired. The infrared images include multiple infrared pixels, and each infrared pixel includes an infrared temperature value.

[0159] For each of the multiple infrared images, perform the following operation:

[0160] Calculate the temperature variance based on multiple infrared pixels in the infrared image;

[0161] Summarize the temperature variances to obtain multiple temperature variances. Calculate the average temperature variance and the advanced temperature standard deviation based on the multiple temperature variances. The average temperature variance is the average of the multiple temperature variances, and the advanced temperature standard deviation is the standard deviation of the multiple temperature variances.

[0162] Infrared volatility is calculated based on the mean temperature variance and the advanced temperature standard deviation, where infrared volatility is the product of the mean temperature variance and the advanced temperature standard deviation.

[0163] The personnel disturbance index is calculated based on infrared volatility, the number of people entering and exiting, and the current number of people. The historical moment when the personnel disturbance index is calculated is recorded.

[0164] By summarizing historical data such as external wind speed, external light intensity, external temperature, set temperature, set wind speed, set brightness, room brightness, room humidity, room temperature, and occupant disturbance index, a historical parameter set is obtained.

[0165] The historical parameter group is integrated into a historical data packet and stored in a pre-built waiting memory to obtain the historical memory. The time of obtaining the historical memory is used as the starting point and the time is recorded in real time to obtain the recording time. When the recording time reaches a preset recording time threshold, the historical memory is used as a waiting memory, and the process of performing the following operations on each of the multiple internal rooms is returned until a pre-built collection end instruction is received, and multiple historical data packets are extracted from the historical memory.

[0166] Historical parameter sets are obtained by using multiple historical data packets, and these sets are then aggregated to obtain multiple historical parameter sets, each corresponding one-to-one with an internal room.

[0167] In detail, the formula for calculating the personnel disturbance index is as follows:

[0168]

[0169] Where, τ P σ represents the personnel disturbance index. T Let N be the infrared fluctuation rate, N0 be the current number of people, and N be the number of people in the current population. x Let ln be the number of people entering and exiting, ln be the natural logarithm, and e be the natural constant.

[0170] It should be explained that external wind speed refers to the wind speed at the initial location outside the initial building; external light intensity refers to the intensity of natural light at the initial location outside the initial building; and external temperature refers to the air temperature at the initial location outside the initial building. Room brightness refers to the light intensity at the location of the brightness sensor inside the room; room humidity refers to the relative humidity of the air at the location of the humidity sensor inside the room; and room temperature refers to the air temperature at the location of the temperature sensor inside the room.

[0171] It should be understood that, historically, during the initial building management process, the wind speed, temperature level, and brightness of the variable-temperature air conditioners and multi-level lighting devices were manually set by the occupants of the corresponding rooms. Therefore, the set temperature and wind speed of the variable-temperature air conditioners and the set brightness of the multi-level lighting devices in the rooms can be historically confirmed. In this embodiment of the invention, to save energy consumption within the initial building and improve the comfort of occupants, multiple historical parameter sets are collected to facilitate the subsequent establishment of multiple energy consumption control models, thereby achieving energy consumption control of the initial building. The set temperature refers to the currently set temperature level of the variable-temperature air conditioner, the set wind speed refers to the currently set wind speed level of the variable-temperature air conditioner, and the set brightness refers to the currently set brightness of the multi-level lighting devices.

[0172] It should be explained that the current number of people refers to the number of people in the internal room at this time, and the number of people entering and exiting refers to the sum of the number of people leaving and entering the internal room within the monitoring time interval. Furthermore, the technology of using a people detector to identify the current number of people in the internal room and the technology of obtaining the number of people entering and exiting the internal room based on a preset monitoring time interval and entry / exit detectors are both existing technologies and will not be elaborated upon here. Optionally, the monitoring time interval is 30 minutes.

[0173] For example, if the shooting time is 1 minute and the shooting interval is 5 seconds, then an infrared image of the interior room is captured by an infrared camera every 5 seconds until the shooting time reaches 1 minute. The captured images are then combined to obtain 12 infrared images. An infrared pixel refers to a pixel in an infrared image.

[0174] Understandably, because the infrared camera receives the infrared radiation energy emitted by people and the internal environment of the room through its internal infrared detector array, converts it into electrical signals, and then restores it to a temperature value to generate an infrared image, each infrared pixel in the infrared temperature image corresponds to an infrared temperature value. This infrared temperature value reflects the temperature of the point corresponding to that infrared pixel in reality. The temperature variance is the variance of the multiple infrared temperature values ​​corresponding to multiple infrared pixels. The infrared volatility reflects the degree of temperature difference between different locations in the room; the greater the infrared volatility, the greater the degree of temperature difference between different locations in the room. The historical moment refers to the moment when the personnel disturbance index is calculated. For example, if the moment when the personnel disturbance index is calculated is 10:00, then the historical moment is 10:00.

[0175] It should be explained that integrating historical parameter groups into historical data packets means storing the parameters in the historical parameter groups as data in the historical data packets. The waiting memory is a memory used to store the historical data packets, and the historical memory is the memory after storing the historical data packets. The recording time threshold is a time interval manually set by the staff of the energy consumption control center; optionally, the recording time threshold is 1 hour.

[0176] For example, historical data packets are stored in a pre-built waiting memory to obtain a historical memory. If the time of obtaining the historical memory is 10:00 on May 1st, then the time is recorded in real time starting from 10:00 on May 1st. When it is 10:06 on May 1st, the recorded time is 6 minutes. If the recording time threshold is 1 hour, then when it is 11:00 on May 1st, the historical memory is used as a waiting memory, and the step of performing the following operation for each of the multiple internal rooms is returned, that is, collecting the historical parameter group at this time again, and continuing to integrate the historical parameter group into historical data packets and store them in the waiting memory, and so on, until the collection end instruction is received. The historical data packets collected in the above cycle process are extracted from the historical memory to obtain multiple historical data packets.

[0177] Understandably, since each historical data packet contains a set of historical parameters, obtaining the set of historical parameters using multiple historical data packets means: sequentially extracting the historical parameter set from each historical data packet and summing the historical parameter sets to obtain the set of historical parameters. The collection termination command is initiated by staff at the energy consumption control center. For example, Xiao Zhang is a staff member at the energy consumption control center. When Xiao Zhang confirms that a certain amount of historical data packets are stored in the historical memory and can be used for subsequent model training, he initiates the collection termination command to end the collection of historical data packets.

[0178] It should be understood that the personnel disturbance index reflects the degree of disturbance to the indoor thermal environment caused by personnel entering, leaving and moving around. The higher the personnel disturbance index, the greater the degree of disturbance to the indoor thermal environment caused by personnel entering, leaving and moving around.

[0179] In detail, the behavioral modeling of multiple pre-constructed sets of historical parameters yields multiple sets of personnel time periods and multiple energy consumption control models, including:

[0180] Perform the following operation for each of the multiple sets of historical parameter sets:

[0181] Construct the overall condition matrix and overall result matrix using historical parameter sets;

[0182] The pre-constructed deep learning model is trained using the conditional total matrix and the result total matrix to obtain the energy consumption regulation model;

[0183] Perform the following operation on each historical parameter group in the historical parameter group set:

[0184] By combining the historical time and personnel disturbance index in the historical parameter group, a personnel time group is obtained;

[0185] Summarize the personnel time groups to obtain a personnel time group set, which includes multiple personnel time groups;

[0186] By summarizing the personnel time sets and energy consumption control models, multiple personnel time sets and multiple energy consumption control models are obtained.

[0187] It should be understood that training a pre-constructed deep learning model using the conditional total matrix and the result total matrix to obtain an energy consumption control model refers to the following: The conditional total matrix is ​​input into the deep learning model, which generates an output matrix based on the input conditional total matrix. Then, the difference between the output matrix and the result total matrix is ​​calculated using a loss function (such as mean squared error). The deep learning model is then corrected based on this difference and the gradient descent algorithm. The conditional total matrix is ​​then input into the corrected deep learning model again. Through continuous iteration of the above process, the deep learning model gradually learns the mapping relationship between the conditional total matrix and the result total matrix. Furthermore, the techniques for calculating the difference between the output matrix and the result total matrix using a loss function and for correcting the deep learning model based on this difference and the gradient descent algorithm are existing technologies and will not be elaborated upon here. The deep learning model is an artificial neural network; preferably, it is a multilayer perceptron (MLP).

[0188] For example, if the historical time is 10:01 and the personnel disturbance index is 2.5, then the personnel time group is: {10:01, 2.5}.

[0189] In detail, the construction of the conditional total matrix and the result total matrix using the historical parameter set includes:

[0190] Multiple external wind speeds, multiple external light intensities, multiple external temperatures, multiple set temperatures, multiple set wind speeds, multiple set brightness, multiple room brightness, multiple room humidity, multiple room temperatures, and multiple occupant disturbance indices were extracted from the historical parameter set.

[0191] Perform the following operation on each historical parameter group in the historical parameter group set:

[0192] The normalized external wind speed is obtained by normalizing the external wind speed in the historical parameter set based on multiple external wind speeds.

[0193] The following parameters are used to obtain the following information: normalized external light intensity, normalized external temperature, normalized set temperature, normalized set wind speed, normalized set wind speed, normalized set brightness, normalized room brightness, normalized room humidity, normalized room humidity, normalized room temperature, and normalized disturbance index.

[0194] A conditional initial matrix is ​​constructed based on the normalized external wind speed, normalized external light intensity, normalized external temperature, normalized design temperature, normalized design wind speed, normalized design brightness, and normalized disturbance index. The conditional initial matrix is ​​shown below:

[0195] Z x =[N wf N wg N wf N sw N sf N sl N r ]

[0196] Among them, Z x Let N be the initial conditional matrix. wf N wg N wf N sw N sf N sl and N r These are, respectively, normalized external wind speed, normalized external light intensity, normalized external temperature, normalized design temperature, normalized design wind speed, normalized design brightness, and normalized disturbance index;

[0197] An initial matrix was constructed based on the normalized room brightness, normalized room humidity, and normalized room temperature. The initial matrix is ​​shown below:

[0198] Z y =[N fl N fs N fw ]

[0199] Among them, Z y Let N be the initial matrix of the result. fl N fs and N fw These are the brightness, humidity, and temperature of the unified room, respectively.

[0200] The initial condition matrix and the initial result matrix are combined to obtain the control training group;

[0201] By summarizing the control training groups, multiple control training groups were obtained;

[0202] A conditional total matrix was constructed based on multiple control training groups, as shown below:

[0203]

[0204] Among them, W X Z is the conditional total matrix. x1 Z is the initial conditional matrix for the first control training group among multiple control training groups. x2 Z is the initial conditional matrix for the second set of multiple control training groups. xi Z is the initial conditional matrix for the i-th control training group among multiple control training groups. xk Let be the initial conditional matrix of the kth control training group among multiple control training groups, where k is the number of control training groups among multiple control training groups;

[0205] A total result matrix was constructed based on multiple control training groups, as shown below:

[0206]

[0207] Among them, W Y Z is the total result matrix. y1 Z is the initial matrix of the results for the first control training group out of multiple control training groups. y2 Z is the initial matrix of the results for the second set of multiple control training groups. yi Z is the initial matrix of the results for the i-th control training group among multiple control training groups. yk This is the initial matrix of the results for the k-th control training group among multiple control training groups.

[0208] In detail, the normalization operation of the external wind speeds in the historical parameter set based on multiple external wind speeds to obtain the normalized external wind speed includes:

[0209] The maximum and minimum wind speeds were determined based on multiple external wind speeds, where the maximum and minimum wind speeds are the largest and smallest external wind speeds among the multiple external wind speeds, respectively.

[0210] The normalized external wind speed is calculated based on the maximum wind speed, minimum wind speed, and external wind speeds from the historical parameter set. The calculation formula is as follows:

[0211]

[0212] Where, N wf For the normalized external wind speed, Nwfx N represents the external wind speed in the historical parameter set. max and N min These are the maximum and minimum wind speeds, respectively.

[0213] Understandably, since the historical parameter set includes multiple historical parameter groups, and each historical parameter group includes historical time, external wind speed, external light intensity, external temperature, set temperature, set wind speed, set brightness, room brightness, room humidity, room temperature, and occupant disturbance index, multiple external wind speeds, multiple external light intensities, multiple external temperatures, multiple set temperatures, multiple set wind speeds, multiple set brightness, multiple room brightness, multiple room humidity, multiple room temperatures, and multiple occupant disturbance indices can be extracted from the multiple historical parameter groups in the historical parameter set.

[0214] It should be understood that the methods for obtaining normalized external light intensity based on multiple external light intensities and historical parameter groups, obtaining normalized external temperature based on multiple external temperatures and historical parameter groups, obtaining normalized set temperature based on multiple set temperatures and historical parameter groups, obtaining normalized set wind speed based on multiple set wind speeds and historical parameter groups, obtaining normalized set brightness based on multiple set brightness and historical parameter groups, obtaining normalized room brightness based on multiple room brightness and historical parameter groups, obtaining normalized room humidity based on multiple room humidity and historical parameter groups, obtaining normalized room temperature based on multiple room temperature and historical parameter groups, and obtaining normalized disturbance index based on multiple personnel disturbance indices and historical parameter groups are all the same as the method for obtaining normalized external wind speed using multiple external wind speeds and historical parameter groups, and will not be repeated here.

[0215] For example, Z x Z is the initial conditional matrix. y As the initial matrix of the results, the control training group is {Z}. x Z y}

[0216] S5. When a pre-built energy consumption control instruction is received, the current time is determined, and multiple predicted disturbance indices are determined in multiple personnel time groups based on the current time.

[0217] It should be explained that the energy consumption control command is initiated by staff of the energy consumption control center. For example, Xiao Zhang is a staff member of the energy consumption control center and now needs to control the energy consumption of the initial building, so he initiates the energy consumption control command. The energy consumption control center is the management department responsible for controlling the operating power consumption of all variable temperature air conditioners and multi-level lighting devices in the initial building. The staff of this department control the energy consumption level of all variable temperature air conditioners and multi-level lighting devices in the initial building by setting the fan speed level, temperature level and luminous brightness of the variable temperature air conditioners in different rooms. Please refer to the following embodiments for the specific control process.

[0218] Specifically, the process of identifying multiple predicted disturbance indices based on the current time across multiple personnel time groups includes:

[0219] The current time range is determined based on the current moment and the preset reference duration;

[0220] Perform the following operation on each of the multiple personnel time groups:

[0221] Based on the current time range, multiple target personnel time groups were identified in the personnel time group set, where the historical times in the target personnel time groups are located within the current time range;

[0222] The predicted disturbance index is calculated based on multiple target personnel time groups, where the predicted disturbance index is the average of the personnel disturbance indices in multiple target personnel time groups;

[0223] By summing the predicted disturbance indices, multiple predicted disturbance indices are obtained, and each predicted disturbance index corresponds one-to-one with a set of personnel time periods.

[0224] It is understandable that "current time" refers to the time at this moment. The minimum value of the current time range is the current time minus the reference duration, and the maximum value of the current time range is the current time plus the reference duration. The reference duration is manually set by the staff of the energy consumption control center. For example, if the reference duration is 30 minutes and the current time is 10:00:00, then the current time range is [09:30:00, 10:30:00].

[0225] It should be explained that the target personnel time group refers to the personnel time group whose historical times are located within the current time range.

[0226] It should be understood that since the working hours of people in the initial building or the usage time of a specific room are usually fixed or regular, the embodiments of the present invention use multiple sets of historical parameters to model behavior, obtain multiple sets of personnel time, and then use multiple sets of personnel time to estimate multiple predicted disturbance indices corresponding to multiple internal rooms at the current time. This makes it easier to input the predicted disturbance indices into the energy consumption control model to predict the optimal machine parameter set for each internal room within a specific time period.

[0227] S6. Use the wind speed sensor, light sensor and external temperature sensor in the external environment detection mechanism to obtain the current wind speed, current light intensity and current external temperature respectively.

[0228] It is understood that the method for obtaining the current wind speed, current light intensity, and current external temperature using the wind speed sensor, light sensor, and external temperature sensor in the external environment detection mechanism is the same as the method for obtaining the external wind speed, external light intensity, and external temperature using the wind speed sensor, light sensor, and external temperature sensor in the external environment detection mechanism, and will not be repeated here. The current wind speed refers to the external wind speed at this moment, the current light intensity refers to the external light intensity at this moment, and the current external temperature refers to the external temperature at this moment.

[0229] S7. Based on multiple optimal condition sets, multiple predicted disturbance indices, current light intensity, current wind speed, and current external temperature, particle swarm simulation is performed on multiple energy consumption control models to obtain multiple optimal machine parameter sets.

[0230] In detail, the process involves particle swarm optimization based on multiple sets of optimal conditions, multiple predicted perturbation indices, current light intensity, current wind speed, and current external temperature to obtain multiple sets of optimal machine parameters, including:

[0231] Confirm the fan speed and temperature selection ranges for the variable temperature air conditioner, and confirm the brightness selection ranges for the multi-level lighting device;

[0232] Based on the wind speed selection range, the first speed selection range is determined, as shown below:

[0233] [-α v ×(v fmax -v fmin ),α v ×(V fmax -V fmin )]

[0234] Where, α v V is the preset velocity coefficient. fmax V is the maximum value of the selected range for wind speed. fmin Select the minimum value within the range for wind speed;

[0235] The second speed selection range is determined based on the temperature selection range, and the third speed selection range is determined based on the brightness selection range.

[0236] For each of the multiple energy consumption control models, perform the following operation:

[0237] The predicted disturbance index corresponding to the energy consumption control model among multiple predicted disturbance indices is denoted as the target disturbance index, and the optimal condition group corresponding to the energy consumption control model among multiple optimal condition groups is denoted as the target optimal condition group.

[0238] Based on multiple room brightness, multiple room humidity, multiple room temperature, and the optimal temperature, optimal humidity, and optimal brightness in the target optimal condition group, obtain the normalized target brightness, normalized target humidity, and normalized target temperature.

[0239] Obtain a particle swarm, wherein the particle swarm includes: multiple particles;

[0240] Perform the following operation on each particle in the particle swarm:

[0241] The wind speed selection range is randomly selected to obtain the position of the wind speed particles;

[0242] The positions of temperature particles, brightness particles, first particle velocities, second particle velocities, and third particle velocities are obtained based on the temperature selection range, brightness selection range, first velocity selection range, second velocity selection range, and third velocity selection range, respectively.

[0243] The three-dimensional positions of the particles were determined based on the positions of wind speed particles, temperature particles, and brightness particles, as shown below:

[0244] P0=(x f ,y w ,z l )

[0245] Where P0 is the three-dimensional position of the particle, x f y w and z l These are the positions of wind speed particles, temperature particles, and brightness particles, respectively.

[0246] The three-dimensional velocities of the particles are determined based on the velocities of the first, second, and third particles, as shown below:

[0247]

[0248] in, The three-dimensional velocities of the particles are v1, v2, and v3, which are the velocities of the first, second, and third particles, respectively.

[0249] The particles are labeled by setting their three-dimensional position and three-dimensional velocity.

[0250] By summing the labeled particles, we obtain a swarm of labeled particles;

[0251] Based on the current light intensity, current wind speed, current external temperature, energy consumption control model, normalized target brightness, normalized target humidity, normalized target temperature and target disturbance index, the labeled particle swarm is iterated to obtain the optimal set of machine parameters.

[0252] By summarizing the optimal machine parameter sets, multiple optimal machine parameter sets are obtained, and each optimal machine parameter set corresponds one-to-one with an energy consumption control model.

[0253] It should be explained that the wind speed selection range is constructed from the minimum and maximum wind speeds that can be set by the variable temperature air conditioner. The minimum value of the wind speed selection range is the minimum wind speed, and the maximum value is the maximum wind speed. The temperature selection range is constructed from the minimum and maximum temperatures that can be set by the variable temperature air conditioner. The minimum value of the temperature selection range is the minimum temperature, and the maximum value is the maximum temperature. The brightness selection range is constructed from the minimum and maximum brightness that can be set by the multi-level lighting device. The minimum value of the brightness selection range is the minimum brightness, and the maximum value is the maximum brightness. Furthermore, the specific minimum and maximum wind speeds, minimum and maximum temperatures are related to the model of the variable temperature air conditioner, and the specific minimum and maximum brightness are related to the model of the multi-level lighting device. The speed coefficient is a value manually set by the staff of the energy consumption control center, and the value range of the speed coefficient is [0.1, 0.2]. The method for determining the second speed selection range based on the temperature selection range and the method for determining the third speed selection range based on the brightness selection range are the same as the method for determining the first speed selection range based on the wind speed selection range, and will not be repeated here.

[0254] It should be understood that the method for obtaining the normalized target brightness, normalized target humidity, and normalized target temperature based on multiple room brightness, multiple room humidity, multiple room temperature, and the optimal temperature, optimal humidity, and optimal brightness in the target optimal condition group is the same as the method for obtaining the normalized room brightness, normalized room humidity, and normalized room temperature using multiple room brightness, multiple room humidity, multiple room temperature, and room brightness, room humidity, and room temperature in the historical parameter group, and will not be repeated here.

[0255] Understandably, a particle swarm optimization algorithm consists of a collection of multiple particles that can move within a three-dimensional search space defined by a range of wind speed, a range of temperature, and a range of brightness. The position of each particle represents a solution within this three-dimensional search space (i.e., a specific wind speed, a specific temperature, and a specific brightness). The three-dimensional search space includes three dimensions: wind speed, temperature, and brightness.

[0256] It should be understood that the method of randomly selecting a wind speed range to obtain the wind speed particle position means randomly selecting a value from the wind speed range as the wind speed particle position. The methods for obtaining temperature particle positions based on a temperature range, brightness particle positions based on a brightness range, first particle velocities based on a first velocity range, second particle velocities based on a second velocity range, and third particle velocities based on a third velocity range are all the same as the method of randomly selecting a wind speed range to obtain the wind speed particle position, and will not be elaborated further here.

[0257] Understandably, the process of setting particles based on their three-dimensional position and three-dimensional velocity to obtain labeled particles means: setting the particle's position in the three-dimensional search space as its three-dimensional position, and setting the particle's movement velocity in the three-dimensional search space as its three-dimensional velocity. A labeled particle swarm is a collection of multiple labeled particles.

[0258] In detail, the step of performing particle swarm iteration on the marked particle swarm based on the current light intensity, current wind speed, current external temperature, energy consumption control model, normalized target brightness, normalized target humidity, normalized target temperature, and target perturbation index to obtain the optimal machine parameter set includes:

[0259] Multiple particle memories were identified, and each particle memory corresponds one-to-one with a labeled particle in the labeled particle swarm.

[0260] For each labeled particle in the labeled particle swarm, perform the following operation:

[0261] The initial particle matrix is ​​obtained based on the current light intensity, current wind speed, current external temperature, target disturbance index, wind speed particle position, temperature particle position, brightness particle position in the three-dimensional position of the marked particle, multiple external wind speeds, multiple external light intensities, multiple external temperatures, multiple set temperatures, multiple set wind speeds, multiple set brightness, and multiple personnel disturbance indices.

[0262] The initial particle matrix is ​​input into the energy consumption control model to obtain the particle result matrix, which includes: predicted room brightness, predicted room humidity and predicted room temperature.

[0263] Particle fitness is calculated based on predicted room brightness, predicted room humidity, predicted room temperature, normalized target brightness, normalized target humidity, and normalized target temperature, using the following formula:

[0264]

[0265] Where, γ z For particle fitness, N ml N ms N mw R L R S and R W These are the normalized target brightness, normalized target humidity, normalized target temperature, predicted room brightness, predicted room humidity, and predicted room temperature, respectively. || represents taking the absolute value.

[0266] The particle fitness and the three-dimensional position of the labeled particle are integrated into a particle data package, and the particle data package is stored in the particle memory corresponding to the particle data package to obtain the update memory.

[0267] The target data packet is identified based on the updated memory, where the target data packet is the particle data packet with the lowest particle fitness among all particle data packets in the updated memory;

[0268] The target data packets and update memories are summarized separately to obtain multiple target data packets and multiple update memories, where each target data packet and update memory corresponds one-to-one with a marked particle.

[0269] The optimal data packet is determined based on multiple target data packets. The optimal data packet is the target data packet with the smallest particle fitness among the multiple target data packets.

[0270] The global optimal position is determined based on the best data packet, where the global optimal position is the three-dimensional position of the particle in the best data packet;

[0271] For each of the multiple target data packets, perform the following operation:

[0272] The labeled particles corresponding to the target data packet are updated based on the three-dimensional position and the global optimal position of the particles in the target data packet, thus obtaining the updated particles;

[0273] Summarize the updated particles to obtain the updated particle swarm;

[0274] The global fitness is obtained based on the wind speed particle position, temperature particle position, brightness particle position, normalized target brightness, normalized target humidity and normalized target temperature in the global optimal position, and the global fitness is compared with the preset fitness threshold.

[0275] If the global fitness is greater than the fitness threshold, the updated particle swarm is used as the labeled particle swarm, the multiple update memories are used as multiple particle memories, and the step of performing the following operation on each labeled particle in the labeled particle swarm is returned until the global fitness is less than or equal to the fitness threshold.

[0276] If the global fitness is less than or equal to the fitness threshold, then the optimal wind speed, optimal temperature and optimal brightness are determined based on the global optimal position, where the optimal wind speed is the wind speed particle position in the global optimal position, the optimal temperature is the temperature particle position in the global optimal position, and the optimal brightness is the brightness particle position in the global optimal position.

[0277] Confirm the multiple settable fan speeds and multiple settable temperatures of the variable temperature air conditioner, and confirm the multiple settable brightness levels of the multi-level lighting device;

[0278] Based on the optimal wind speed, the nearest optimal wind speed is identified among multiple settable wind speeds; based on the optimal temperature, the nearest optimal temperature is identified among multiple settable temperature ranges; based on the optimal brightness, the nearest optimal brightness is identified among multiple settable brightness ranges.

[0279] The optimal set of machine parameters is obtained by summarizing the nearest optimal wind speed, nearest optimal temperature, and nearest optimal brightness.

[0280] It should be explained that a particle memory is a memory used to store particle data packets. Integrating particle fitness and 3D particle positions into a particle data packet means storing particle fitness and 3D particle positions as data in the data packet to obtain the particle data packet. The update memory is the particle memory that stores the particle data packet.

[0281] It is understood that the method for obtaining the initial particle matrix based on the current light intensity, current wind speed, current external temperature, target disturbance index, wind speed particle position, temperature particle position, brightness particle position, multiple external wind speeds, multiple external light intensities, multiple external temperatures, multiple set temperatures, multiple set wind speeds, multiple set brightness, and multiple personnel disturbance indices in the three-dimensional position of the marked particles is the same as the method for obtaining the initial condition matrix using external light intensity, external wind speed, external temperature, personnel disturbance index, set wind speed, set temperature, set brightness, multiple external wind speeds, multiple external light intensities, multiple external temperatures, multiple set temperatures, multiple set wind speeds, multiple set brightness, and multiple personnel disturbance indices in the historical parameter group, and will not be described in detail here.

[0282] It should be understood that since the energy consumption control model is trained using the initial condition matrix and the initial result matrix, when the initial particle matrix is ​​input into the energy consumption control model, the energy consumption control model will output a matrix. The elements in the output matrix have a one-to-one correspondence with the elements in the initial result matrix. That is, the element in the first row and first column of the output matrix represents brightness, which is the predicted room brightness; the element in the first row and second column represents humidity, which is the predicted room humidity; and the element in the first row and third column represents temperature, which is the predicted room temperature.

[0283] Understandably, during the first iteration, the update memory contains only one particle data packet, which serves as the target data packet. In subsequent iterations, the number of particle data packets in the update memory increases, and the target data packet becomes the one with the lowest particle fitness among all the particle data packets in the update memory. It's important to note that when the update particle swarm is used as the marker particle swarm, each update particle in the update particle swarm is correspondingly converted into a marker particle, and the updated 3D position and updated 3D velocity of the update particles are also converted into the 3D position and initial velocity of the marker particles, respectively, to ensure the iterative progress of the loop steps.

[0284] Understandably, the adaptation threshold is a value set manually by the staff of the energy consumption control center, and the range of the adaptation threshold is

[10] . -6 10 -3 This ensures that the differences between the predicted room brightness, predicted room humidity, and predicted room temperature and the normalized target brightness, normalized target humidity, and normalized target temperature are small enough, so that after setting the parameters of multiple internal rooms in the initial building according to multiple optimal machine parameter groups, the brightness, humidity, and temperature in the internal rooms are closer to the optimal brightness, optimal humidity, and optimal temperature, respectively.

[0285] It should be understood that the method for obtaining global fitness based on the wind speed particle position, temperature particle position, brightness particle position, normalized target brightness, normalized target humidity, and normalized target temperature in the global optimal position is the same as the method for obtaining particle fitness using the wind speed particle position, temperature particle position, brightness particle position, normalized target brightness, normalized target humidity, and normalized target temperature in the three-dimensional position of the marked particle, and will not be repeated here.

[0286] For example, since variable temperature air conditioners cannot be set to any temperature, but can only switch between multiple specific temperature levels, such as 16℃, 17℃, ..., 30℃, these multiple specific temperature levels are multiple settable temperatures. Similarly, variable temperature air conditioners can only switch between multiple specific wind speed levels, which are multiple settable wind speeds. And multi-level lighting devices can only switch between multiple specific brightness levels, which are multiple settable brightness levels. The specific multiple settable wind speeds and multiple settable temperatures are related to the model of the variable temperature air conditioner, and the specific multiple settable brightness levels are related to the model of the multi-level lighting device.

[0287] It is understood that the determination of the nearest optimal wind speed among multiple settable wind speeds based on the optimal wind speed means: determining the settable wind speed with the smallest absolute difference from the optimal wind speed among multiple settable wind speeds, and taking the settable wind speed with the smallest absolute difference as the nearest optimal wind speed.

[0288] It should be understood that the method for determining the nearest optimal temperature from multiple settable temperatures based on the optimal temperature and the method for determining the nearest optimal brightness from multiple settable brightness based on the optimal brightness are the same as the method for determining the nearest optimal wind speed from multiple settable wind speeds based on the optimal wind speed, and will not be described again here.

[0289] Specifically, updating the marked particles corresponding to the target data packet based on the three-dimensional position and the global optimal position of the particles in the target data packet to obtain updated particles includes:

[0290] The preset unit value range is randomly extracted twice to obtain the first random number and the second random number;

[0291] The updated wind speed is calculated based on the first random number, the second random number, the wind speed particle position in the three-dimensional position of the target data packet, the wind speed particle position in the global optimal position, the first particle velocity in the three-dimensional velocity of the marked particle, and the wind speed particle position in the three-dimensional position of the marked particle. The calculation formula is as follows:

[0292] v fn =ω0×v b1 +β1×(x pf -x f0 )+β2×(x qf -x f0 )

[0293] Among them, v fn To update the wind speed, v0 is the initial particle velocity of the marker particle, x pf and x qfThese represent the wind speed particle positions in the three-dimensional particle positions of the target data packet and the wind speed particle positions in the global optimal position, respectively. f0 To mark the position of the wind speed particle in the three-dimensional position of the particle, ω0 is the preset inertial weight, and β1 and β2 are the first random number and the second random number, respectively;

[0294] The updated wind speed position is calculated based on the wind speed particle position in the three-dimensional position of the marked particle and the updated wind speed velocity. The calculation formula is as follows:

[0295] x fgx =x f0 +v fn

[0296] Where, x fgx To update wind speed location;

[0297] The updated temperature position and updated temperature velocity are obtained based on the first random number, the second random number, the temperature particle position in the three-dimensional position of the target data packet, the temperature particle position in the global optimal position, the second particle velocity in the three-dimensional velocity of the marked particle, and the temperature particle position in the three-dimensional position of the marked particle.

[0298] The updated brightness position and updated brightness velocity are obtained based on the first random number, the second random number, the brightness particle position in the three-dimensional position of the target data packet, the brightness particle position in the global optimal position, the third particle velocity in the three-dimensional velocity of the marked particle, and the brightness particle position in the three-dimensional position of the marked particle.

[0299] The updated three-dimensional position is determined based on the updated wind speed position, updated temperature position, and updated brightness position; the updated three-dimensional velocity is determined based on the updated wind speed velocity, updated temperature velocity, and updated brightness velocity.

[0300] The updated particles are obtained by setting the marker particles based on the updated three-dimensional position and updated three-dimensional velocity.

[0301] It should be understood that the unit value range is [0,1]. The phrase "two random extractions from the preset unit value range to obtain the first random number and the second random number" means: randomly selecting a value from the unit value range as the first random number, and then randomly selecting a value from the unit value range as the second random number. The inertia weight is a value manually set by the staff of the energy consumption control center, and the inertia weight range is [0.6, 0.9].

[0302] It is understood that the methods for obtaining updated temperature positions and updated temperature velocities based on the first random number, the second random number, the temperature particle positions in the three-dimensional positions of the target data packet, the temperature particle positions in the globally optimal positions, the second particle velocity in the three-dimensional velocities of the marked particles, and the temperature particle positions in the three-dimensional positions of the marked particles, as well as the methods for obtaining updated brightness positions and updated brightness velocities based on the first random number, the second random number, the brightness particle positions in the three-dimensional positions of the target data packet, the brightness particle positions in the globally optimal positions, the third particle velocity in the three-dimensional velocities of the marked particles, and the brightness particle positions in the three-dimensional positions of the marked particles, are all the same as the methods for obtaining updated wind speed positions and updated wind speed velocities using the first random number, the second random number, the wind speed particle positions in the three-dimensional positions of the target data packet, the wind speed particle positions in the globally optimal positions, the first particle velocity in the three-dimensional velocities of the marked particles, and the wind speed particle positions in the three-dimensional positions of the marked particles, and will not be described in detail here.

[0303] It should be understood that the method for determining the updated three-dimensional position based on the updated wind speed position, updated temperature position, and updated brightness position is the same as the method for determining the particle's three-dimensional position based on the wind speed particle position, temperature particle position, and brightness particle position, and will not be repeated here. Similarly, the method for determining the updated three-dimensional velocity based on the updated wind speed, updated temperature speed, and updated brightness speed is the same as the method for determining the particle's three-dimensional velocity based on the first particle velocity, second particle velocity, and third particle velocity, and will not be repeated here. The method for setting marker particles based on the updated three-dimensional position and updated three-dimensional velocity to obtain updated particles is the same as the method for setting particles based on the particle's three-dimensional position and particle's three-dimensional velocity to obtain marked particles, and will not be repeated here.

[0304] S8. Using multiple optimal machine parameter sets, set parameters for multiple internal rooms in the initial building to obtain the target building. Starting from the time of obtaining the target building, record the time in real time to obtain the stage time. Until the stage time reaches the preset stage threshold, use the target building as the initial building and return to the step of confirming the current time. Continue until a pre-constructed control end command is received to complete the energy consumption control of the building.

[0305] For example, after obtaining multiple optimal machine parameter groups, the following operations are performed sequentially on each of the multiple optimal machine parameter groups: the wind speed of the variable temperature air conditioner in the internal room corresponding to the optimal machine parameter group is set to the nearest optimal wind speed in the optimal machine parameter group, the temperature of the variable temperature air conditioner is set to the nearest optimal temperature in the optimal machine parameter group, and the brightness of the multi-level lighting device in the internal room is set to the nearest optimal brightness in the optimal machine parameter group, until the multiple variable temperature air conditioners and multiple multi-level lighting devices in the multiple internal rooms are all set, and the initial building at this time is taken as the target building. If the time of the target building is 10:00:00, then 10:00:00 is taken as the starting point and the time is recorded in real time to obtain the stage time. When 10:02:00, the stage time is 2 minutes. If the stage time threshold is 1 hour, then at 11:00:00, the target building is taken as the initial building, and the process returns to the step of confirming the current time. That is, multiple optimal machine parameter sets corresponding to the time of 11:00:00 are obtained again, so as to adjust the energy consumption of the initial building in real time and in stages until the control end command is received, and the energy consumption control of the initial building ends.

[0306] Understandably, the control termination command is initiated by staff at the energy consumption control center. The command is sent when it is necessary to terminate energy consumption control of the initial building. The stage threshold is manually set by staff at the energy consumption control center; optionally, the stage threshold is 1 hour.

[0307] It should be understood that the embodiments of the present invention construct a cyclical step through phased time, thereby obtaining multiple optimal machine parameter sets that are most suitable for the current initial building in stages, and setting the variable temperature air conditioning and multi-level lighting devices of multiple internal rooms according to the multiple optimal machine parameter sets, thereby saving energy consumption and automatically providing the most suitable internal environment for the people inside the initial building at different times.

[0308] To address the problems described in the background art, this invention identifies an initial building, comprising multiple internal rooms, each including a multi-source sensing mechanism, a variable-temperature air conditioner, and multi-level lighting. The multi-source sensing mechanism includes a people detector, an access detector, an infrared camera, a brightness sensor, a humidity sensor, and a temperature sensor. This invention, by identifying the initial building, clarifies the deployment of various sensing devices within the multiple internal rooms, providing data support and a spatial basis for the implementation of subsequent energy consumption control strategies. Furthermore, it identifies an external environment detection mechanism, including a wind speed sensor, a light sensor, and an external temperature sensor. This invention, by identifying the external environment detection mechanism, thereby achieves… Dynamic perception of the building's external environment facilitates subsequent particle swarm simulation based on current wind speed, light intensity, and external temperature. This allows for the assessment of suitable conditions in multiple internal rooms, resulting in multiple optimal condition sets. Each optimal condition set corresponds one-to-one with an internal room and includes optimal temperature, optimal humidity, and optimal brightness. Therefore, this embodiment of the invention, by assessing suitable conditions in multiple internal rooms, clarifies the optimal environmental control target among them, facilitating subsequent particle swarm simulation with specific objectives based on the optimal condition sets. This improves the comfort of occupants within the building. Furthermore, behavioral modeling is performed on pre-constructed sets of historical parameters, resulting in multiple sets of occupant timeframes and multiple energy consumption control models. These historical parameter sets and occupant timeframes are used to model behavior, leading to multiple sets of occupant timeframes and multiple energy consumption control models. Each personnel time set and energy consumption control model corresponds one-to-one with an internal room. This invention demonstrates that by modeling the behavior of multiple historical parameter sets, multiple personnel time sets are obtained, reflecting the activity patterns of personnel within the initial building at different time periods. This leads to the construction of multiple energy consumption control models, which predict the environmental state of internal rooms under specific external environments, set temperatures, set wind speeds, and set brightness levels. This facilitates subsequent particle swarm simulation based on the energy consumption control models. When a pre-constructed energy consumption control command is received, the current time is determined. Based on this current time, multiple predicted disturbance indices are identified from the multiple personnel time sets. Wind speed sensors, light sensors, and external temperature sensors in the external environment detection mechanism are used to obtain... The current wind speed, current light intensity, and current external temperature demonstrate that this embodiment of the invention provides real-time external parameters for subsequent energy consumption control by confirming multiple predicted disturbance indices and current external environmental data. This improves the environmental suitability and accuracy of energy consumption control. Based on multiple optimal condition sets, multiple predicted disturbance indices, current light intensity, current wind speed, and current external temperature, particle swarm simulation is performed on multiple energy consumption control models to obtain multiple optimal machine parameter sets. This embodiment of the invention inputs multi-source data into the particle swarm algorithm for simulation calculation. By continuously trying multiple different temperature, wind speed, and brightness settings and substituting them into the energy consumption control model to calculate particle fitness, the optimal machine parameter set that best matches the optimal condition set is globally searched.This invention utilizes multiple optimal machine parameter sets to set parameters for multiple internal rooms in an initial building, resulting in a target building. Starting from the time the target building is obtained, the time is recorded in real-time to obtain a stage time. This stage time is repeated until a preset stage threshold is reached. The target building is then used as the initial building, and the process returns to the step of confirming the current time. This continues until a pre-constructed control end command is received, completing the energy consumption control of the building. Thus, this embodiment of the invention sequentially sets the variable-temperature air conditioning and multi-level lighting devices in each internal room of the initial building using multiple optimal machine parameter sets, thereby improving the comfort of occupants. Through a cyclical step of stage time construction, new optimal machine parameter sets are regenerated stage by stage based on the current environmental state, achieving staged adaptive optimization control. This effectively avoids energy efficiency losses caused by redundant energy output under mismatched environmental conditions, thereby reducing building energy consumption. Therefore, this invention can improve the comfort of occupants and reduce building energy consumption.

[0309] like Figure 2 The diagram shown is a functional block diagram of an intelligent building energy consumption control system based on multi-source data fusion provided in an embodiment of the present invention.

[0310] The intelligent building energy consumption control system 100 based on multi-source data fusion described in this invention can be installed in an electronic device 1. Depending on the functions implemented, the intelligent building energy consumption control system 100 based on multi-source data fusion may include an initial environment confirmation module 101, a historical parameter modeling module 102, a building parameter optimization module 103, and an energy consumption control cycle module 104. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0311] The initial environment confirmation module 101 is used to confirm the initial building, wherein the initial building includes: multiple internal rooms, and the internal rooms include: multi-source sensing mechanism, variable temperature air conditioner and multi-level lighting device. The multi-source sensing mechanism includes: people recognition device, entry and exit detector, infrared camera, brightness sensor, humidity sensor and temperature sensor. The module also confirms the external environment detection mechanism, wherein the external environment detection mechanism includes: wind speed sensor, light sensor and external temperature sensor.

[0312] The historical parameter modeling module 102 is used to evaluate the suitability of multiple internal rooms and obtain multiple optimal condition groups. The optimal condition groups correspond one-to-one with the internal rooms and include: optimal temperature, optimal humidity and optimal brightness. The module performs behavioral modeling on the pre-constructed sets of historical parameter groups to obtain multiple sets of personnel time and multiple energy consumption control models. The historical parameter groups, personnel time groups and energy consumption control models all correspond one-to-one with the internal rooms.

[0313] The building parameter optimization module 103 is used to determine the current time when it receives a pre-constructed energy consumption control instruction, determine multiple predicted disturbance indices based on the current time in multiple personnel time groups, use the wind speed sensor, light sensor and external temperature sensor in the external environment detection mechanism to obtain the current wind speed, current light intensity and current external temperature respectively, and perform particle swarm simulation on multiple energy consumption control models based on multiple optimal condition groups, multiple predicted disturbance indices, current light intensity, current wind speed and current external temperature to obtain multiple optimal machine parameter groups;

[0314] The energy consumption control loop module 104 is used to set parameters for multiple internal rooms in the initial building using multiple optimal machine parameter sets to obtain the target building. Starting from the time of obtaining the target building, the time is recorded in real time to obtain the stage time. Until the stage time reaches the preset stage threshold, the target building is used as the initial building, and the process returns to the step of confirming the current time. The process continues until a pre-constructed control end command is received to complete the energy consumption control of the building.

[0315] In detail, the modules in the intelligent building energy consumption control system 100 based on multi-source data fusion described in this embodiment of the invention employ the same methods as described above during use. Figure 1 The method used is the same as the intelligent building energy consumption control method based on multi-source data fusion described in the article, and can produce the same technical effect, so it will not be repeated here.

[0316] like Figure 3 The diagram shown is a structural schematic of an electronic device 1 that implements a smart building energy consumption control method based on multi-source data fusion, according to an embodiment of the present invention.

[0317] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and capable of running on the processor 10, such as a program for intelligent building energy consumption control based on multi-source data fusion.

[0318] The memory 11 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as a portable hard drive. In other embodiments, the memory 11 can be an external storage device of the electronic device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device 1. Furthermore, the memory 11 includes both internal storage units and external storage devices of the electronic device 1. The memory 11 can be used not only to store application software and various types of data installed on the electronic device 1, such as the code of a smart building energy consumption control method program based on multi-source data fusion, but also to temporarily store data that has been output or will be output.

[0319] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., a smart building energy consumption control method program based on multi-source data fusion) and calls data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0320] The bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0321] Figure 3 Only electronic devices with components are shown; those skilled in the art will understand that... Figure 3The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0322] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0323] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device 1 and other electronic devices.

[0324] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired or wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual user interface.

[0325] The program for intelligent building energy consumption control based on multi-source data fusion, stored in the memory 11 of the electronic device 1, is a combination of multiple instructions. When run in the processor 10, it can achieve the following:

[0326] The initial building has been identified, which includes: multiple internal rooms, and each internal room includes: a multi-source sensing mechanism, a variable temperature air conditioner, and a multi-level lighting device. The multi-source sensing mechanism includes: a people recognition device, an access detector, an infrared camera, a brightness sensor, a humidity sensor, and a temperature sensor.

[0327] The external environment detection mechanisms have been identified, including: wind speed sensor, light sensor and external temperature sensor;

[0328] The suitability conditions of multiple internal rooms were evaluated to obtain multiple optimal condition groups. Each optimal condition group corresponds to an internal room and includes: optimal temperature, optimal humidity, and optimal brightness.

[0329] Behavioral modeling is performed on multiple pre-constructed sets of historical parameters to obtain multiple sets of personnel time and multiple energy consumption control models. The sets of historical parameters, personnel time, and energy consumption control models are all one-to-one correspondences with the internal rooms.

[0330] When a pre-built energy consumption control instruction is received, the current time is determined, and multiple predicted disturbance indices are determined in multiple personnel time groups based on the current time.

[0331] The wind speed sensor, light sensor and external temperature sensor in the external environment detection mechanism are used to obtain the current wind speed, current light intensity and current external temperature respectively;

[0332] Based on multiple sets of optimal conditions, multiple predicted disturbance indices, current light intensity, current wind speed, and current external temperature, particle swarm simulation was performed on multiple energy consumption control models to obtain multiple sets of optimal machine parameters.

[0333] The system uses multiple optimal machine parameter sets to set parameters for multiple internal rooms in the initial building to obtain the target building. Starting from the time when the target building is obtained, the system records the time in real time to obtain the stage time. The stage time is continued until the stage time reaches the preset stage threshold. The target building is then used as the initial building, and the system returns to the step of confirming the current time. This process continues until a pre-constructed control end command is received, thus completing the energy consumption control of the building.

[0334] Specifically, the processor 10's implementation method for the above instructions can be found in [reference needed]. Figures 1 to 3 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0335] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0336] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0337] The initial building has been identified, which includes: multiple internal rooms, and each internal room includes: a multi-source sensing mechanism, a variable temperature air conditioner, and a multi-level lighting device. The multi-source sensing mechanism includes: a people recognition device, an access detector, an infrared camera, a brightness sensor, a humidity sensor, and a temperature sensor.

[0338] The external environment detection mechanisms have been identified, including: wind speed sensor, light sensor and external temperature sensor;

[0339] The suitability conditions of multiple internal rooms were evaluated to obtain multiple optimal condition groups. Each optimal condition group corresponds to an internal room and includes: optimal temperature, optimal humidity, and optimal brightness.

[0340] Behavioral modeling is performed on multiple pre-constructed sets of historical parameters to obtain multiple sets of personnel time and multiple energy consumption control models. The sets of historical parameters, personnel time, and energy consumption control models are all one-to-one correspondences with the internal rooms.

[0341] When a pre-built energy consumption control instruction is received, the current time is determined, and multiple predicted disturbance indices are determined in multiple personnel time groups based on the current time.

[0342] The wind speed sensor, light sensor and external temperature sensor in the external environment detection mechanism are used to obtain the current wind speed, current light intensity and current external temperature respectively;

[0343] Based on multiple sets of optimal conditions, multiple predicted disturbance indices, current light intensity, current wind speed, and current external temperature, particle swarm simulation was performed on multiple energy consumption control models to obtain multiple sets of optimal machine parameters.

[0344] The system uses multiple optimal machine parameter sets to set parameters for multiple internal rooms in the initial building to obtain the target building. Starting from the time when the target building is obtained, the system records the time in real time to obtain the stage time. The stage time is continued until the stage time reaches the preset stage threshold. The target building is then used as the initial building, and the system returns to the step of confirming the current time. This process continues until a pre-constructed control end command is received, thus completing the energy consumption control of the building.

[0345] In the embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and actual implementations may have other classification methods.

[0346] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0347] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0348] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0349] Finally, 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.

Claims

1. A method for intelligent building energy consumption control based on multi-source data fusion, characterized in that, The method includes: The initial building has been identified, which includes: multiple internal rooms, and each internal room includes: a multi-source sensing mechanism, a variable temperature air conditioner, and a multi-level lighting device. The multi-source sensing mechanism includes: a people recognition device, an access detector, an infrared camera, a brightness sensor, a humidity sensor, and a temperature sensor. The external environment detection mechanisms have been identified, including: wind speed sensor, light sensor and external temperature sensor; The suitability conditions of multiple internal rooms were evaluated to obtain multiple optimal condition groups. Each optimal condition group corresponds to an internal room and includes: optimal temperature, optimal humidity, and optimal brightness. Behavioral modeling is performed on multiple pre-constructed sets of historical parameters to obtain multiple sets of personnel time and multiple energy consumption control models. The sets of historical parameters, personnel time, and energy consumption control models are all one-to-one correspondences with the internal rooms. When a pre-built energy consumption control instruction is received, the current time is determined, and multiple predicted disturbance indices are determined in multiple personnel time groups based on the current time. The wind speed sensor, light sensor and external temperature sensor in the external environment detection mechanism are used to obtain the current wind speed, current light intensity and current external temperature respectively; Based on multiple sets of optimal conditions, multiple predicted disturbance indices, current light intensity, current wind speed, and current external temperature, particle swarm simulation was performed on multiple energy consumption control models to obtain multiple sets of optimal machine parameters. The system uses multiple optimal machine parameter sets to set parameters for multiple internal rooms in the initial building to obtain the target building. Starting from the time when the target building is obtained, the system records the time in real time to obtain the stage time. The stage time is continued until the stage time reaches the preset stage threshold. The target building is then used as the initial building, and the system returns to the step of confirming the current time. This process continues until a pre-constructed control end command is received, thus completing the energy consumption control of the building.

2. The intelligent building energy consumption control method based on multi-source data fusion as described in claim 1, characterized in that, The assessment of suitability conditions for multiple internal rooms yields multiple sets of optimal conditions, including: Perform the following operation on each of the multiple internal rooms: Obtain the historical assessment dataset of the internal rooms. The historical assessment dataset includes multiple historical assessment data points, including assessment temperature, assessment brightness, assessment humidity, and assessment scores. The historical assessment dataset is sorted in descending order of assessment scores to obtain a historical assessment data sequence. Confirm the number of assessments for multiple historical assessment data points in the historical assessment dataset; The target number is determined based on the preset optimal ratio and the number of evaluations, where the target number is the product of the optimal ratio and the number of evaluations. Based on the target quantity, multiple optimal evaluation data are extracted from the historical evaluation data sequence; Multiple evaluation temperatures, multiple evaluation brightness, and multiple evaluation humidity were extracted from multiple optimal evaluation data; The optimal temperature is calculated based on multiple measured temperatures, where the optimal temperature is the average of the multiple measured temperatures. The optimal brightness is calculated based on multiple brightness measurements, and the optimal humidity is calculated based on multiple humidity measurements. The optimal conditions are obtained by combining the optimal temperature, optimal brightness, and optimal humidity. By summarizing the optimal condition groups, multiple optimal condition groups are obtained.

3. The intelligent building energy consumption control method based on multi-source data fusion as described in claim 2, characterized in that, Before performing behavioral modeling on multiple pre-constructed sets of historical parameters to obtain multiple sets of personnel time intervals and multiple energy consumption control models, the process also includes: Perform the following operation on each of the multiple internal rooms: The external wind speed, external light intensity, and external temperature are obtained by using the wind speed sensor, light sensor, and external temperature sensor in the external environment detection mechanism, respectively. The brightness, humidity and temperature of the interior room are collected by the brightness sensor, humidity sensor and temperature sensor in the multi-source sensing mechanism, respectively. Confirm the set temperature and fan speed of the variable temperature air conditioner in the room, and confirm the set brightness of the multi-level lighting device; Use a people detector to identify the current number of people in the rooms; The number of people entering and exiting the internal rooms is obtained based on the preset monitoring time interval and the access detectors. Based on a preset shooting time, shooting time interval, and infrared camera, multiple infrared images of the interior room are acquired. The infrared images include multiple infrared pixels, and each infrared pixel includes an infrared temperature value. For each of the multiple infrared images, perform the following operation: Calculate the temperature variance based on multiple infrared pixels in the infrared image; Summarize the temperature variances to obtain multiple temperature variances. Calculate the average temperature variance and the advanced temperature standard deviation based on the multiple temperature variances. The average temperature variance is the average of the multiple temperature variances, and the advanced temperature standard deviation is the standard deviation of the multiple temperature variances. Infrared volatility is calculated based on the mean temperature variance and the advanced temperature standard deviation, where infrared volatility is the product of the mean temperature variance and the advanced temperature standard deviation. The personnel disturbance index is calculated based on infrared volatility, the number of people entering and exiting, and the current number of people. The historical moment when the personnel disturbance index is calculated is recorded. By summarizing historical data such as external wind speed, external light intensity, external temperature, set temperature, set wind speed, set brightness, room brightness, room humidity, room temperature, and occupant disturbance index, a historical parameter set is obtained. The historical parameter group is integrated into a historical data packet and stored in a pre-built waiting memory to obtain the historical memory. The time of obtaining the historical memory is used as the starting point and the time is recorded in real time to obtain the recording time. When the recording time reaches a preset recording time threshold, the historical memory is used as a waiting memory, and the process of performing the following operations on each of the multiple internal rooms is returned until a pre-built collection end instruction is received, and multiple historical data packets are extracted from the historical memory. Historical parameter sets are obtained by using multiple historical data packets, and these sets are then aggregated to obtain multiple historical parameter sets, each corresponding one-to-one with an internal room.

4. The intelligent building energy consumption control method based on multi-source data fusion as described in claim 3, characterized in that, The formula for calculating the personnel disturbance index is as follows: Where, τ P σ represents the personnel disturbance index. T Let N be the infrared fluctuation rate, N0 be the current number of people, and N be the number of people in the current population. x Let ln be the number of people entering and exiting, ln be the natural logarithm, and e be the natural constant.

5. The intelligent building energy consumption control method based on multi-source data fusion as described in claim 4, characterized in that, The process involves performing behavioral modeling on multiple pre-constructed sets of historical parameters to obtain multiple sets of personnel time points and multiple energy consumption control models, including: Perform the following operation for each of the multiple sets of historical parameter sets: Construct the overall condition matrix and overall result matrix using historical parameter sets; The pre-constructed deep learning model is trained using the conditional total matrix and the result total matrix to obtain the energy consumption regulation model; Perform the following operation on each historical parameter group in the historical parameter group set: By combining the historical time and personnel disturbance index in the historical parameter group, a personnel time group is obtained; Summarize the personnel time groups to obtain a personnel time group set, which includes multiple personnel time groups; By summarizing the personnel time sets and energy consumption control models, multiple personnel time sets and multiple energy consumption control models are obtained.

6. The intelligent building energy consumption control method based on multi-source data fusion as described in claim 5, characterized in that, The construction of the conditional total matrix and the result total matrix using historical parameter sets includes: Multiple external wind speeds, multiple external light intensities, multiple external temperatures, multiple set temperatures, multiple set wind speeds, multiple set brightness, multiple room brightness, multiple room humidity, multiple room temperatures, and multiple occupant disturbance indices were extracted from the historical parameter set. Perform the following operation on each historical parameter group in the historical parameter group set: The normalized external wind speed is obtained by normalizing the external wind speed in the historical parameter set based on multiple external wind speeds. The following parameters are used to obtain the following information: normalized external light intensity, normalized external temperature, normalized set temperature, normalized set wind speed, normalized set wind speed, normalized set brightness, normalized room brightness, normalized room humidity, normalized room humidity, normalized room temperature, and normalized disturbance index. Construct an initial condition matrix based on normalized external wind speed, normalized external light intensity, normalized external temperature, normalized design temperature, normalized design wind speed, normalized design brightness, and normalized disturbance index; An initial matrix of results is constructed based on the brightness, humidity, and temperature of the unified room. The initial condition matrix and the initial result matrix are combined to obtain the control training group; By summarizing the control training groups, multiple control training groups were obtained; Construct a conditional total matrix based on multiple control training groups; A total result matrix was constructed based on multiple control training groups.

7. The intelligent building energy consumption control method based on multi-source data fusion as described in claim 6, characterized in that, The method identifies multiple predicted disturbance indices based on the current time across multiple personnel time groups, including: The current time range is determined based on the current moment and the preset reference duration; Perform the following operation on each of the multiple personnel time groups: Based on the current time range, multiple target personnel time groups were identified in the personnel time group set, where the historical times in the target personnel time groups are located within the current time range; The predicted disturbance index is calculated based on multiple target personnel time groups, where the predicted disturbance index is the average of the personnel disturbance indices in multiple target personnel time groups; By summing the predicted disturbance indices, multiple predicted disturbance indices are obtained, and each predicted disturbance index corresponds one-to-one with a set of personnel time periods.

8. The intelligent building energy consumption control method based on multi-source data fusion as described in claim 7, characterized in that, The process involves particle swarm optimization (PSO) simulations of multiple energy consumption control models based on multiple sets of optimal conditions, multiple predicted disturbance indices, current light intensity, current wind speed, and current external temperature to obtain multiple sets of optimal machine parameters, including: Confirm the fan speed and temperature selection ranges for the variable temperature air conditioner, and confirm the brightness selection ranges for the multi-level lighting device; The first speed selection range is determined based on the wind speed selection range; The second speed selection range is determined based on the temperature selection range, and the third speed selection range is determined based on the brightness selection range. For each of the multiple energy consumption control models, perform the following operation: The predicted disturbance index corresponding to the energy consumption control model among multiple predicted disturbance indices is denoted as the target disturbance index, and the optimal condition group corresponding to the energy consumption control model among multiple optimal condition groups is denoted as the target optimal condition group. Based on multiple room brightness, multiple room humidity, multiple room temperature, and the optimal temperature, optimal humidity, and optimal brightness in the target optimal condition group, obtain the normalized target brightness, normalized target humidity, and normalized target temperature. Obtain a particle swarm, wherein the particle swarm includes: multiple particles; Perform the following operation on each particle in the particle swarm: The wind speed selection range is randomly selected to obtain the position of the wind speed particles; The positions of temperature particles, brightness particles, first particle velocities, second particle velocities, and third particle velocities are obtained based on the temperature selection range, brightness selection range, first velocity selection range, second velocity selection range, and third velocity selection range, respectively. The three-dimensional position of the particles was determined based on the positions of wind speed particles, temperature particles, and brightness particles. The three-dimensional velocity of the particle is determined based on the velocities of the first, second, and third particles. The particles are labeled by setting their three-dimensional position and three-dimensional velocity. By summing the labeled particles, we obtain a swarm of labeled particles; Based on the current light intensity, current wind speed, current external temperature, energy consumption control model, normalized target brightness, normalized target humidity, normalized target temperature and target disturbance index, the labeled particle swarm is iterated to obtain the optimal set of machine parameters. By summarizing the optimal machine parameter sets, multiple optimal machine parameter sets are obtained, and each optimal machine parameter set corresponds one-to-one with an energy consumption control model.

9. The intelligent building energy consumption control method based on multi-source data fusion as described in claim 8, characterized in that, The process involves iterating through the labeled particle swarm based on current light intensity, current wind speed, current external temperature, energy consumption control model, normalized target brightness, normalized target humidity, normalized target temperature, and target perturbation index to obtain the optimal set of machine parameters, including: Multiple particle memories were identified, and each particle memory corresponds one-to-one with a labeled particle in the labeled particle swarm. For each labeled particle in the labeled particle swarm, perform the following operation: The initial particle matrix is ​​obtained based on the current light intensity, current wind speed, current external temperature, target disturbance index, wind speed particle position, temperature particle position, brightness particle position in the three-dimensional position of the marked particle, multiple external wind speeds, multiple external light intensities, multiple external temperatures, multiple set temperatures, multiple set wind speeds, multiple set brightness, and multiple personnel disturbance indices. The initial particle matrix is ​​input into the energy consumption control model to obtain the particle result matrix, which includes: predicted room brightness, predicted room humidity and predicted room temperature. Particle fitness is calculated based on predicted room brightness, predicted room humidity, predicted room temperature, normalized target brightness, normalized target humidity, and normalized target temperature, using the following formula: Where, γ z For particle fitness, N ml N ms N mw R L R S and R W These are the normalized target brightness, normalized target humidity, normalized target temperature, predicted room brightness, predicted room humidity, and predicted room temperature, respectively. || represents taking the absolute value. The particle fitness and the three-dimensional position of the labeled particle are integrated into a particle data package, and the particle data package is stored in the particle memory corresponding to the particle data package to obtain the update memory. The target data packet is identified based on the updated memory, where the target data packet is the particle data packet with the lowest particle fitness among all particle data packets in the updated memory; The target data packets and update memories are summarized separately to obtain multiple target data packets and multiple update memories, where each target data packet and update memory corresponds one-to-one with a marked particle. The optimal data packet is determined based on multiple target data packets. The optimal data packet is the target data packet with the smallest particle fitness among the multiple target data packets. The global optimal position is determined based on the best data packet, where the global optimal position is the three-dimensional position of the particle in the best data packet; Perform the following operation on each of the multiple target data packets: The labeled particles corresponding to the target data packet are updated based on the three-dimensional position and the global optimal position of the particles in the target data packet, thus obtaining the updated particles; Summarize the updated particles to obtain the updated particle swarm; The global fitness is obtained based on the wind speed particle position, temperature particle position, brightness particle position, normalized target brightness, normalized target humidity and normalized target temperature in the global optimal position, and the global fitness is compared with the preset fitness threshold. If the global fitness is greater than the fitness threshold, the updated particle swarm is used as the labeled particle swarm, the multiple update memories are used as multiple particle memories, and the step of performing the following operation on each labeled particle in the labeled particle swarm is returned until the global fitness is less than or equal to the fitness threshold. If the global fitness is less than or equal to the fitness threshold, then the optimal wind speed, optimal temperature and optimal brightness are determined based on the global optimal position, where the optimal wind speed is the wind speed particle position in the global optimal position, the optimal temperature is the temperature particle position in the global optimal position, and the optimal brightness is the brightness particle position in the global optimal position. Confirm the multiple settable fan speeds and multiple settable temperatures of the variable temperature air conditioner, and confirm the multiple settable brightness levels of the multi-level lighting device; Based on the optimal wind speed, the nearest optimal wind speed is identified among multiple settable wind speeds; based on the optimal temperature, the nearest optimal temperature is identified among multiple settable temperature ranges; based on the optimal brightness, the nearest optimal brightness is identified among multiple settable brightness ranges. The optimal set of machine parameters is obtained by summarizing the nearest optimal wind speed, nearest optimal temperature, and nearest optimal brightness.

10. An intelligent building energy consumption control system based on multi-source data fusion, characterized in that, The system includes: The initial environment confirmation module is used to confirm the initial building, which includes multiple internal rooms, and each internal room includes a multi-source sensing mechanism, a variable temperature air conditioner, and a multi-level lighting device. The multi-source sensing mechanism includes a people recognition device, an access detector, an infrared camera, a brightness sensor, a humidity sensor, and a temperature sensor. The module also confirms the external environment detection mechanism, which includes a wind speed sensor, a light sensor, and an external temperature sensor. The historical parameter modeling module is used to evaluate the suitability of multiple internal rooms and obtain multiple optimal condition groups. Each optimal condition group corresponds one-to-one with an internal room and includes optimal temperature, optimal humidity, and optimal brightness. The module performs behavioral modeling on the pre-constructed sets of historical parameter groups to obtain multiple sets of personnel time and multiple energy consumption control models. Each set of historical parameter groups, personnel time groups, and energy consumption control models corresponds one-to-one with an internal room. The building parameter optimization module is used to determine the current time when a pre-built energy consumption control command is received. Based on the current time, it determines multiple predicted disturbance indices from multiple personnel time groups. It uses wind speed sensors, light sensors, and external temperature sensors in the external environment detection mechanism to obtain the current wind speed, current light intensity, and current external temperature, respectively. Based on multiple optimal condition groups, multiple predicted disturbance indices, current light intensity, current wind speed, and current external temperature, it performs particle swarm simulation on multiple energy consumption control models to obtain multiple optimal machine parameter groups. The energy consumption control loop module is used to set parameters for multiple internal rooms in the initial building using multiple optimal machine parameter sets to obtain the target building. Starting from the time of obtaining the target building, the time is recorded in real time to obtain the stage time. Until the stage time reaches the preset stage threshold, the target building is used as the initial building, and the process returns to the step of confirming the current time. The process continues until a pre-constructed control end command is received, thus completing the energy consumption control of the building.

Images