An intelligent building energy consumption optimization method, device and medium based on a cloud platform

Abstract

The present application discloses an intelligent building energy consumption optimization method, device and medium based on a cloud platform. The method includes: obtaining the historical data of the intelligent building and the historical working parameters of the energy consumption equipment in the previous preset time period; inputting them into a large power industry model to obtain the first predicted indoor data for the current preset time period; obtaining the actual indoor data and the current working parameters; obtaining the prediction error and selecting an optimal power industry model; obtaining the future indoor data for the next preset time period and calculating the corresponding average value; determining the building working conditions for the next preset time period and matching an adapted industry model; establishing an energy consumption optimization objective function for solving the minimum energy consumption of the intelligent building on the condition that the indoor environment value is within the preset comfort range; obtaining the minimum energy consumption and determining the working parameters of the energy consumption equipment. Through the unified management of the large model by the cloud platform, not only can the accuracy and reliability of the model be improved, but also the cost of data acquisition and processing can be reduced.

Description

Technical Field

[0001] This application relates to the technical field of intelligent building energy management, and particularly relates to an intelligent building energy consumption optimization method, device, and medium based on a cloud platform. Background Art

[0002] With the rapid development of Internet of Things, artificial intelligence, and big data analysis technologies and their applications in buildings, more and more buildings are designed and constructed as intelligent buildings. Along with the increasing global energy consumption and the growing prominence of environmental problems, the attention of governments, enterprises, and individuals to building energy efficiency has been continuously enhanced, and the energy consumption management of intelligent buildings has been increasingly emphasized.

[0003] Existing energy consumption management methods usually require deploying large model algorithms locally in intelligent buildings and using building data in the ontology intelligent building to train the constructed large models. The energy consumption of intelligent buildings is predicted through the trained models. The large models constructed by this method can only be applied in local intelligent buildings. And when upgrading the server, the model upgrade cost is relatively high, the robustness is poor, and in the face of system fluctuations or changes, its stability and adaptability are insufficient. The model data comes from a single building and is difficult to be applied to building groups, resulting in waste of model resources. Summary of the Invention

[0004] To solve the above problems, this application proposes an intelligent building energy consumption optimization method based on a cloud platform, which is applied in the cloud platform and includes:

[0005] For the current time node, obtain the historical data of the intelligent building and the historical working parameters of the energy consumption equipment within the previous preset time period; the historical data includes historical indoor data and historical outdoor data; the historical indoor data includes historical indoor environmental data and historical energy consumption data, and the historical outdoor data includes historical outdoor environmental data;

[0006] Input the historical data and the historical working parameters into each deployed large model in the power industry to obtain the first predicted indoor data of each large model in the power industry for the current preset time period;

[0007] Obtain the actual indoor data of the intelligent building and the current working parameters of the energy consumption equipment within the current preset time period;

[0008] According to the first predicted indoor data and the actual indoor data, obtain the prediction error of each large model in the power industry, and select the preferred large models in the power industry whose prediction errors are lower than the preset threshold;

[0009] Input the actual indoor data and the current working parameters into the preferred power industry large model to obtain the future indoor data for the next preset time period, and calculate the average values corresponding to the future indoor data respectively;

[0010] Based on the average values, determine the building conditions for the next preset time period, and match the adapted industry large model corresponding to the building conditions in the preferred power industry large model;

[0011] According to the second predicted indoor data output by the adapted industry large model for the next preset time period, with the condition that the indoor environment value is within the preset comfortable range, establish an energy consumption optimization objective function for solving the minimum value of the intelligent building energy consumption;

[0012] Solve the optimal solution of the energy consumption optimization objective function to obtain the minimum energy consumption value, and determine the working parameters of the energy consumption equipment according to the minimum energy consumption value to adjust the energy consumption equipment.

[0013] On the other hand, the present application also proposes an intelligent building energy consumption optimization device based on a cloud platform, including:

[0014] At least one processor; and,

[0015] A memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute an intelligent building energy consumption optimization method based on a cloud platform as described in the above example.

[0017] On the other hand, the present application also proposes a non-volatile computer storage medium storing computer-executable instructions, and the computer-executable instructions are set as: an intelligent building energy consumption optimization method based on a cloud platform as described in the above example.

[0018] By the intelligent building energy consumption optimization method based on a cloud platform proposed by the present application, the following

[0019] Beneficial effects can be brought:

[0020] Through the powerful computing power of the cloud platform, a large amount of data collected in the intelligent building can be centrally processed, so that the prediction of energy consumption data is more accurate. Multiple different power industry large models are pre-deployed in the cloud platform, the large models are uniformly managed through the cloud platform, and the large models share the intelligent building data stored in the cloud platform through the data interface provided by the cloud platform, which can not only improve the accuracy and reliability of the models, but also reduce the cost of data acquisition and processing.

[0021] The average value obtained through the prediction of multiple large models in the power industry is used to determine the building conditions according to the average value, and the corresponding large model in the industry is determined according to the conditions, making the model more suitable for intelligent buildings, so that the obtained prediction data is more accurate. Brief Description of the Drawings

[0022] The drawings described herein are used to provide a further understanding of the present application, form a part of the present application, and the schematic embodiments and descriptions thereof are used to explain the present application, and do not constitute an improper limitation to the present application. In the drawings:

[0023] Figure 1 It is a schematic flow chart of an intelligent building energy consumption optimization method based on a cloud platform in an embodiment of the present application;

[0024] Figure 2 It is a schematic diagram of the relationship between the large model in the power industry and the intelligent building in an embodiment of the present application;

[0025] Figure 3 It is a schematic diagram of a variable edge computing power topology structure in an embodiment of the present application;

[0026] Figure 4 It is a schematic diagram of an intelligent building energy consumption optimization device based on a cloud platform in an embodiment of the present application. Detailed Description of the Embodiments

[0027] To make the objectives, technical solutions, and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below in conjunction with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present application.

[0028] The following will detail the technical solutions provided by each embodiment of the present application in conjunction with the drawings.

[0029] It should be noted that the large model in the industry is a machine learning model customized and optimized for a specific field or industry. Compared with the general large model, the large model in the industry can make full use of its large parameter scale in a specific field, capture complex correlations in the data, and has stronger professionalism. With the development of artificial intelligence technology, the large model in the industry has been more widely applied in various fields and has achieved remarkable results in many fields. The large model in the industry is a deep learning model that targets the needs of a specific industry or field and uses large-scale data training and advanced algorithms. The large model in the industry usually includes the key processes, value chains, participants, relationships, and other important elements of the industry, which can help understand the operation mechanism, business process, and value creation method of the industry.

[0030] As Figure 1 shown, an embodiment of the present application provides an intelligent building energy consumption optimization method based on a cloud platform, including:

[0031] S101: For the current time node, obtain the historical data of the intelligent building and the historical working parameters of the energy-consuming equipment in the previous preset time period; the historical data includes historical indoor data and historical outdoor data; the historical indoor data includes historical indoor environmental data and historical energy consumption data, and the historical outdoor data includes historical outdoor environmental data.

[0032] Specifically, receive an energy consumption optimization request of the intelligent building, and determine whether the energy consumption optimization request is the first energy consumption optimization request. If so, for the current time node corresponding to the received energy consumption optimization request, obtain the historical data of the intelligent building and the historical working parameters of the energy-consuming equipment in the intelligent building in the previous preset time period. Among them, the historical data includes historical indoor data and historical outdoor data, the historical indoor data includes historical indoor environmental data and historical energy consumption data, and the historical outdoor data includes historical outdoor environmental data.

[0033] If not, execute steps S106 and S107.

[0034] In the embodiment of the specification of the present application, sensor devices are installed both inside and outside the intelligent building. Through the installed sensor devices, indoor environmental data and outdoor environmental data are collected in real time within a preset time period. At the same time, for the collection time node, the working parameters of the energy-consuming equipment inside the intelligent building are obtained to obtain corresponding time series data. After the collection is completed within the preset time period, the energy consumption data of the intelligent building in the preset time period is counted.

[0035] Among them, the indoor environmental data includes parameters such as indoor temperature and humidity, occupancy rate of personnel, etc., and the outdoor environmental data includes parameters such as weather factors. After collecting the indoor environmental data, outdoor environmental data, equipment working parameters, and energy consumption data, a data preprocessing process is performed, including data cleaning, removing duplicate values, filling missing values, etc. After data preprocessing, it is uploaded to the cloud platform for storage through encrypted transmission, and corresponding data interfaces are generated according to different preset time periods. In the subsequent process, if it is necessary to use the collected intelligent building data, the corresponding intelligent building data can be obtained through the data interfaces corresponding to different preset time periods. The equipment working parameters are the operating states of various energy-consuming equipment in the building, such as the on / off, operating mode (cooling / heating / ventilation), current set temperature, wind speed level, etc. of the air conditioning system; the on / off, brightness level, light sensor state (whether to automatically adjust the brightness according to the surrounding light), etc. of the lighting system.

[0036] It should be noted that in the embodiment of the present application, a process for determining the installation positions of the sensor devices in the intelligent building is also included.

[0037] Specifically, according to the building use of the intelligent building (such as office, commercial, residential, industrial, etc., and the corresponding building types are office buildings, commercial buildings, residential buildings, industrial buildings, etc.), determine the operation mode of the intelligent building, analyze the operation mode, and determine the energy consumption mode of the intelligent building. For example, when the building use is for office, its corresponding operation mode is the office mode. Then, weekdays are peak energy consumption time periods, weekends are low peak energy consumption time periods, the floors where large enterprises are located are high energy consumption areas, and small companies with fewer personnel belong to low energy consumption areas, etc.

[0038] Furthermore, based on the energy consumption mode, determine the high energy consumption areas and low energy consumption areas in the intelligent building, set the high energy consumption areas as key locations, and install sensor devices in the key locations, such as key components of heating, ventilation, air conditioning, lighting and other systems, as well as areas with dense personnel or equipment.

[0039] It should be noted that by accurately collecting data at key energy consumption points, the collected data can be accurate and targeted, and energy consumption anomalies and waste situations can be discovered in a timely manner, so that energy-saving management and optimization measures can be quickly implemented to improve energy-saving efficiency. Among them, the energy consumption equipment in the building includes electrical equipment (such as generators, motors, etc.), thermal equipment (such as boilers, heat exchangers, etc.), cooling systems, lighting equipment, and transportation equipment (such as pumps, fans), which use different energy consumptions. The electrical energy consumption of electrical equipment can be expressed as the product of power (P) and time (t), that is, E = P * t, and the thermal energy consumption of thermal equipment can be expressed as the product of the fuel consumption and its thermal efficiency.

[0040] S102: Input the historical data and the historical working parameters into each deployed power industry large model to obtain the first predicted indoor data of each power industry large model for the current preset time period.

[0041] Specifically, use the historical indoor environment data, historical energy consumption data, historical outdoor environment data, and historical working parameters of the intelligent building in the previous preset time period as input data, and input them into each deployed power industry large model in the cloud platform. Through each power industry large model, predict the indoor environment data and energy consumption data for the current preset time period, and respectively obtain the first predicted indoor data of each power industry large model for the current preset time period, including the predicted indoor environment data and predicted energy consumption data for the current preset time period.

[0042] It should be noted that as Figure 2As shown, different large models for the power industry are large models for multiple large model service providers, which are pre-deployed in the cloud platform. Specifically, the large model for the power industry is uploaded to a specified storage location on the cloud platform, and the model is adapted to ensure that it can run normally in the computing environment of the cloud platform, and relevant interfaces of the large model for the power industry are configured in the cloud platform to enable direct invocation of the model. The large model for the power industry is a model that has been trained for predicting the energy consumption data of intelligent buildings. The model architecture includes an input layer, a deep neural network layer, and an output layer. It is relatively mature for energy consumption prediction, has a relatively complete process, and has achieved remarkable application effects.

[0043] In addition, in the embodiments of the present application specification, before inputting the historical data of the intelligent building into the large model for the power industry, the data volume of different large models for the power industry for the intelligent building is counted, so as to determine the computing volume of different large models for the power industry for predicting the intelligent building.

[0044] Specifically, the indoor data and outdoor data of each intelligent building in different time intervals are obtained, including temperature, humidity, light intensity, personnel flow, as well as weather conditions, ambient temperature, etc. According to the indoor data, the computing volume of different large models for the power industry of each intelligent building is determined, and a corresponding computing volume time curve is constructed to represent the change of the computing resource amount required by each intelligent building at different time points.

[0045] More specifically, the indoor data and outdoor data are cleaned and feature extraction is performed. According to the data scale after feature extraction, the input data size of each intelligent building at different time points is evaluated. According to the type, parameter configuration and input data scale of the model, the computing complexity of the model on each intelligent building is evaluated. By simulating the operation of the model or a computing volume evaluation tool, combined with the input data scale and model complexity, the computing volume required by each intelligent building at different time points is estimated. The computing volumes of each intelligent building at different time points are aggregated, and through a curve fitting method (such as linear regression, polynomial regression, etc.), the aggregated computing volume data is fitted into a computing volume time curve.

[0046] Furthermore, based on the computing volume time curve, the peak value, valley value and overall trend of the curve of each intelligent building in different time periods are determined. According to the curve change, the change of the computing volume in different time periods is determined. For example, during the time periods with frequent personnel activities (such as morning and evening), the computing volume may increase, while during the time periods with less personnel activities (such as late at night), the computing volume may decrease.

[0047] Calculate the average or maximum value of the computing volume within this time period according to the change situation, so as to determine the computing power required in each time period, and construct a computing power demand curve according to the computing power. According to the computing power demand curve, determine the minimum participation number of each edge data center through the edge time-varying topology structure algorithm. Based on the determined minimum participation number, construct the edge computing power topology structure of the edge data center. As Figure 3 shown, describe the connection relationship, data transmission path and computing power allocation strategy between edge data centers through the computing power topology structure to ensure that the computing power requirements of intelligent buildings can be met efficiently and flexibly.

[0048] In one or more embodiments of the present application specification, extract the computing power values of the edge data center in different time intervals from the computing power demand curve. According to the computing power values in different time intervals, with the condition that the computing duration required for the computing volume in a single time interval is not higher than the preset duration threshold, and the hardware bearing capacity in a single time interval is not lower than the preset bearing threshold (for example, the predicted historical data volume per hour by the model is not higher than 0.5 seconds, and the cpu load rate is not lower than 75%), establish an optimization objective function for solving the shortest computing response time and the least number of participating edge data centers, and solve the optimal solution of the optimization objective function through the particle swarm algorithm to obtain the minimum participation number corresponding to different times of the edge data center.

[0049] It should be noted that the cloud platform includes multiple edge data centers, and each edge data center manages multiple intelligent buildings or intelligent building groups. Through the intelligent green building energy consumption optimization platform with variable edge computing power topology structure and reasonable deployment of compute-intensive services and latency-sensitive services, it has efficient energy consumption data integration, mining, sharing capabilities and distributed, hierarchical massive data processing capabilities. The average response time of intelligent building cloud services has a significant improvement effect compared with the traditional centralized cloud service system based on large data centers.

[0050] S103: Obtain the actual indoor data of the intelligent building and the current working parameters of the energy consumption equipment within the current preset time period.

[0051] Specifically, after the duration of the preset time period, the sensor collects the intelligent building data of the current preset time period and uploads the intelligent building data to the cloud platform. The cloud platform obtains the actual indoor environment data, actual energy consumption data of the intelligent building and the current working parameters of the energy consumption equipment within the current preset time period.

[0052] S104: According to the first predicted indoor data and the actual indoor data, respectively obtain the prediction errors of each power industry large model, and select the preferred power industry large models with prediction errors lower than the preset threshold.

[0053] Specifically, obtain the prediction data output by the large model of the power industry for the current preset time period, including the prediction of indoor environment data and energy consumption data. Obtain the difference between the prediction data output by multiple large models of the power industry and the actual data, which is the prediction error of the large model of the power industry. Compare multiple prediction errors to determine the maximum prediction error and the corresponding large model of the power industry. Delete the large model of the power industry corresponding to the maximum prediction error to obtain the remaining preferred large models of the power industry.

[0054] It should be noted that in the embodiments of the present application, the prediction error of different large models of the power industry is represented by a loss function. According to the algorithms and structures adopted by the large models of the industry, the type of the loss function to be used is determined, including cross-entropy loss function, mean square error loss function, etc. The loss function used in the embodiments of the present application is the mean square error loss function. Use the loss function calculation formula: where n is the number of samples, y i is the historical energy consumption data of the i-th sample, y i ^ is the predicted energy consumption data of the i-th sample. Based on the historical energy consumption data, determine the prediction error between the predicted energy consumption data and the historical energy consumption data. According to the prediction error and the extreme formula of the loss function, calculate the loss functions of multiple large models of the industry respectively. Compare the values corresponding to the loss functions to obtain the loss function value with the largest value and the corresponding large model of the power industry. Delete the large model of the power industry.

[0055] S104: Input the actual indoor data and the current working parameters into the preferred large model of the power industry to obtain the future indoor data for the next preset time period, and calculate the average values corresponding to the future indoor data respectively.

[0056] Specifically, input the current actual indoor environment data, the current working parameters of energy consumption equipment, and the current actual energy consumption data in the intelligent building during the current preset time period into multiple preferred large models of the power industry. Through the multiple preferred large models of the power industry, output the future indoor environment data and future energy consumption data for the next preset time period, and calculate the first average value of the future indoor environment data and the second average value of the first future energy consumption data respectively.

[0057] In one or more embodiments of the present application, obtain the first future indoor environment data and the first future energy consumption data, and perform a rationality analysis on the first future indoor environment data based on a physical mechanism; when the first future indoor environment data is reasonable, determine whether the first future energy consumption data is within a preset standard range; if not, determine that the first future energy consumption data is abnormal data, and determine the abnormal type corresponding to the abnormal data through a preset classification algorithm to process according to the abnormal type; if so, determine that the first future energy consumption data is normal.

[0058] Among them, the abnormal types include sudden anomalies, periodic anomalies, static anomalies, noise anomalies, etc. A sudden anomaly refers to a sudden sharp increase or decrease in energy consumption within a short period, which may be related to equipment failures, accidents, or human factors. A periodic anomaly usually refers to an anomaly that occurs within a specific time period (such as weekly or monthly), which may be related to regular maintenance activities or operational changes. A static anomaly refers to a high or low energy consumption state that remains unchanged for a long time, which may indicate equipment aging or reduced efficiency. A noise anomaly refers to random fluctuations in the data, which may be caused by sensor failures or data entry errors.

[0059] Specifically, the process of determining the abnormal type includes: evenly dividing the next preset time period into several time intervals, obtaining the first future energy consumption data corresponding to the time intervals, and determining the degree of energy consumption change of the several time intervals based on the first future energy consumption data corresponding to the time intervals. When there is a degree of energy consumption change higher than the preset threshold, determine the corresponding abnormal time interval, and determine whether the abnormal time interval is a random time. If so, determine that the abnormal type corresponding to the abnormal data is a noise anomaly. If not, determine whether the interval length of the time interval is higher than the preset time length. If not, determine that the abnormal type corresponding to the abnormal data is a sudden anomaly. If so, determine that the abnormal type corresponding to the abnormal data is a periodic anomaly. When there is no degree of energy consumption change higher than the preset threshold, determine that the abnormal type corresponding to the abnormal data is a static anomaly.

[0060] S105: Based on the average value, determine the building condition of the next preset time period, and match the adapted industry large model corresponding to the building condition in the preferred power industry large model.

[0061] Specifically, according to the first average value of the first future indoor environment data, determine the indoor environment condition of the intelligent building in the next preset time period. According to the second average value of the first future energy consumption data, determine the energy consumption condition of the intelligent building in the next preset time period. Based on the indoor environment condition and the energy consumption condition, determine the building condition of the next preset time period.

[0062] Furthermore, extract the key feature vectors in the building condition. Based on the key feature vectors, use the matching algorithm in the model library to determine the adapted industry large model corresponding to the building condition. For example, the K-Nearest Neighbors (KNN) algorithm. In the intelligent building scenario, represent the features of the building condition as vectors, and use the KNN algorithm to search for the K neighbors in the model library that are most similar to the target condition, and determine the most matching model according to the labels of these neighbors.

[0063] S106: Based on the second predicted indoor data output by the adapted industry large model for the next preset time period, with the condition that the indoor environmental value is within the preset comfort interval, establish an energy consumption optimization objective function for solving the minimum value of the intelligent building's energy consumption.

[0064] Specifically, based on the second predicted indoor data output by the adapted industry large model for the next preset time period, the energy consumption index is determined as the optimization objective. With the condition that the indoor environmental value (such as temperature, humidity, etc.) must be maintained within the preset comfort interval, an energy consumption optimization objective function is established to minimize the energy consumption of the intelligent building while meeting the indoor comfort requirements.

[0065] It should be noted that in the embodiments of this application, the formula of the objective function is where f(x) is the objective function, E i (x) is the energy consumption of the i-th device, x is the variable parameter, and n is the total number of devices.

[0066] S107: Solve the optimal solution of the energy consumption optimization objective function to obtain the minimum energy consumption value. Determine the working parameters of the energy consumption devices according to the minimum energy consumption value to regulate the energy consumption devices.

[0067] Specifically, according to the model parameters of the adapted industry large model and the optimization objective function, construct an optimization model. Through a personalized optimization algorithm, adjust the optimization variables to achieve the adjustment of the model parameters. By adjusting the variable parameters in the energy consumption objective function and solving the adjusted energy consumption objective function, obtain the corresponding objective results. Set the minimum value in the objective results as the optimal result and determine the variable parameters corresponding to the optimal result. Among them, the personalized optimization algorithms include the particle swarm algorithm and the chimpanzee algorithm.

[0068] Furthermore, determine whether the value corresponding to the optimal result meets the constraint conditions. If not, delete the value corresponding to the optimal result, set the second smallest value in the objective results as the new optimal result, obtain the variable parameters corresponding to the new optimal result, and check whether the new optimal result meets the constraint conditions.

[0069] When receiving a non-first energy consumption optimization request for the intelligent building, based on the current time node of the non-first energy consumption optimization request, obtain the historical data of the intelligent building and the historical working parameters of the energy consumption devices within the corresponding previous preset time period. Input the historical data and historical working parameters corresponding to the non-first energy consumption optimization request into the adapted power industry large model to obtain the predicted indoor data of the adapted power industry large model for the prediction time period. According to the predicted indoor data, through the energy consumption optimization objective function, obtain the minimum energy consumption value corresponding to the non-first energy consumption optimization request.

[0070] With the powerful computing power of the cloud platform, a large amount of data collected in the intelligent building can be centrally processed, making the prediction of energy consumption data more accurate. Multiple different large models in the power industry are pre-deployed in the cloud platform. The cloud platform uniformly manages the large models, and enables the large models to share the intelligent building data stored in the cloud platform through the data interfaces provided by the cloud platform, which can not only improve the accuracy and reliability of the models, but also reduce the costs of data acquisition and processing.

[0071] The average value predicted by multiple large models in the power industry is used to determine the building conditions according to the average value, and the corresponding large model in the industry is determined according to the conditions, making the model more suitable for the intelligent building, and thus making the obtained prediction data more accurate.

[0072] As Figure 4 shown, the embodiment of the present application also proposes an intelligent building energy consumption optimization device based on a cloud platform, including:

[0073] At least one processor; and,

[0074] A memory communicatively connected to the at least one processor; wherein,

[0075] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute a method for optimizing the energy consumption of an intelligent building based on a cloud platform as described in any one of the above embodiments.

[0076] The embodiment of the present application also provides a non-volatile computer storage medium storing computer-executable instructions, and the computer-executable instructions are set to: a method for optimizing the energy consumption of an intelligent building based on a cloud platform as described in any one of the above embodiments.

[0077] The embodiments in the present application are all described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments. In particular, for the embodiments of the device and the medium, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0078] The device and the medium provided by the embodiment of the present application correspond one by one to the method. Therefore, the device and the medium also have beneficial technical effects similar to those of the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the device and the medium will not be elaborated here.

[0079] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0080] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0081] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0082] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are performed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0083] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0084] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM), and / or non-volatile memory such as read-only memory (ROM) or flash memory (flash RAM). The memory is an example of computer-readable media.

[0085] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0086] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0087] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. An intelligent building energy consumption optimization method based on a cloud platform, characterized in that, Applied in a cloud platform, the method includes: For the current time node, obtain the historical data of the intelligent building and the historical working parameters of the energy-consuming equipment in the previous preset time period; the historical data includes historical indoor data and historical outdoor data; the historical indoor data includes historical indoor environmental data and historical energy consumption data, and the historical outdoor data includes historical outdoor environmental data; Input the historical data and the historical working parameters into each deployed large power industry model to obtain the first predicted indoor data of each large power industry model for the current preset time period; Obtain the actual indoor data of the intelligent building and the current working parameters of the energy-consuming equipment in the current preset time period; According to the first predicted indoor data and the actual indoor data, obtain the prediction errors of each large power industry model respectively, and select multiple preferred large power industry models with prediction errors lower than the preset threshold; Input the actual indoor data and the current working parameters into the preferred large power industry models to obtain the future indoor data for the next preset time period, and calculate the average values corresponding to the future indoor data respectively; Based on the average values, determine the building conditions for the next preset time period, and match the appropriate large industry model corresponding to the building conditions in the model library; The determining the building conditions for the next preset time period based on the average values specifically includes: Determine the indoor environmental conditions of the intelligent building in the next preset time period according to the first average value of the future indoor environmental data; Determine the energy consumption conditions of the intelligent building in the next preset time period according to the second average value of the future energy consumption data; Based on the indoor environmental conditions and the energy consumption conditions, determine the building conditions for the next preset time period; According to the second predicted indoor data output by the appropriate large industry model for the next preset time period, and with the condition that the indoor environmental value is within the preset comfortable range, establish an energy consumption optimization objective function for solving the minimum energy consumption of the intelligent building; Solve the optimal solution of the energy consumption optimization objective function to obtain the minimum energy consumption, and determine the working parameters of the energy-consuming equipment according to the minimum energy consumption to regulate the energy-consuming equipment.

2. The intelligent building energy consumption optimization method based on a cloud platform according to claim 1, characterized in that The cloud platform includes multiple edge data centers. After obtaining the historical data of the intelligent building and the historical working parameters of the energy-consuming equipment in the previous preset time period for the current time node, the method further includes: Obtain the indoor data and outdoor data of each intelligent building in different time intervals; Determine the calculation amounts of different large power industry models of each intelligent building according to the indoor data, and construct the corresponding calculation amount time curve; Determine the computing power demand curve of each intelligent building according to the calculation amount time curve; According to the computing power demand curve, determine the minimum participation numbers of each edge data center through the edge time-varying topology structure algorithm, and construct the edge computing power topology structure of the edge data centers based on the minimum participation numbers, and perform energy consumption optimization prediction on each intelligent building according to the edge computing power topology structure.

3. The intelligent building energy consumption optimization method based on a cloud platform according to claim 2, characterized in that, According to the computing power demand curve, determining the minimum participation quantity of each edge data center through the edge time-varying topology structure algorithm specifically includes: Extracting the computing power values of the edge data centers in different time intervals from the computing power demand curve; Based on the computing power values in different time intervals, with the condition that the computing duration required for the computing amount in a single time interval is not higher than a preset duration threshold and the hardware carrying capacity in a single time interval is not lower than a preset carrying threshold, establishing an optimization objective function for solving the shortest computing response time and the least number of participating edge data centers; Solving the optimal solution of the optimization objective function through a particle swarm algorithm to obtain the minimum participation quantity corresponding to different times of the edge data centers.

4. An intelligent building energy consumption optimization method based on a cloud platform according to claim 1, characterized in that, According to the minimum energy consumption value, determining the working parameters of the energy consumption equipment specifically includes: Determining the environmental data value corresponding to the energy consumption optimization objective function at the minimum energy consumption value; Determining the working parameters of the energy consumption equipment according to the environmental data value.

5. The intelligent building energy consumption optimization method based on a cloud platform according to claim 1, characterized in that The method further includes: Determining to receive a non-first energy consumption optimization request of the intelligent building; Based on the non-first energy consumption optimization request, obtaining the historical data of the intelligent building and the historical working parameters of the energy consumption equipment in the corresponding previous preset time period; Inputting the historical data and historical working parameters corresponding to the non-first energy consumption optimization request into an adapted power industry large model to obtain the predicted indoor data of the adapted power industry large model for the prediction time period; According to the predicted indoor data, through the energy consumption optimization objective function, obtaining the minimum energy consumption value corresponding to the non-first energy consumption optimization request.

6. The intelligent building energy consumption optimization method based on a cloud platform according to claim 1, characterized in that, After inputting the actual indoor data and the current working parameters into the preferred power industry large model to obtain the future indoor data of the next preset time period, the method further includes: Performing a rationality analysis on the future indoor environment data in the future indoor data based on a physical mechanism; When the future indoor environment data is reasonable, determining whether the future energy consumption data in the future indoor data is within a preset standard range; If not, determining that the future energy consumption data is abnormal data, and determining the abnormal type corresponding to the abnormal data through a preset classification algorithm to perform processing according to the abnormal type.

7. An intelligent building energy consumption optimization method based on a cloud platform according to claim 6, characterized in that The abnormal types include sudden abnormality, periodic abnormality, static abnormality, and noise abnormality; Determining the abnormal type corresponding to the abnormal data through a preset classification algorithm specifically includes: Evenly dividing the next preset time period into several time intervals, obtaining the interval first future energy consumption data corresponding to the time intervals, and determining the energy consumption change degree of the several time intervals based on the interval first future energy consumption data; When there is an energy consumption change degree higher than a preset threshold, determining the corresponding abnormal time interval and determining whether the abnormal time interval is a random time; If so, determining that the abnormal type corresponding to the abnormal data is noise abnormality; If not, determining whether the interval length of the time interval is higher than a preset time length; If not, determining that the abnormal type corresponding to the abnormal data is sudden abnormality; If so, determine that the anomaly type corresponding to the abnormal data is a periodic anomaly; When there is no degree of energy consumption change higher than the preset threshold, determine that the anomaly type corresponding to the abnormal data is a static anomaly.

8. An intelligent building energy consumption optimization device based on a cloud platform, characterized in that, Including: At least one processor; And, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute as follows: For the current time node, obtain the historical data of the intelligent building and the historical working parameters of the energy consumption equipment in the previous preset time period; the historical data includes historical indoor data and historical outdoor data; the historical indoor data includes historical indoor environmental data and historical energy consumption data, and the historical outdoor data includes historical outdoor environmental data; Input the historical data and the historical working parameters into each deployed large power industry model to obtain the first predicted indoor data of each large power industry model for the current preset time period; Obtain the actual indoor data of the intelligent building and the current working parameters of the energy consumption equipment in the current preset time period; According to the first predicted indoor data and the actual indoor data, respectively obtain the prediction errors of each large power industry model, and select multiple preferred large power industry models with prediction errors lower than the preset threshold; Input the actual indoor data and the current working parameters into the preferred large power industry models to obtain the future indoor data for the next preset time period, and calculate the average values corresponding to the future indoor data respectively; Based on the average values, determine the building condition for the next preset time period, and match the adapted large industry model corresponding to the building condition in the model library; Based on the average values, determining the building condition for the next preset time period specifically includes: According to the first average value of the future indoor environmental data, determine the indoor environmental condition of the intelligent building in the next preset time period; According to the second average value of the future energy consumption data, determine the energy consumption condition of the intelligent building in the next preset time period; Based on the indoor environmental condition and the energy consumption condition, determine the building condition for the next preset time period; According to the second predicted indoor data output by the adapted large industry model for the next preset time period, with the condition that the indoor environmental value is within the preset comfort range, establish an energy consumption optimization objective function for solving the minimum energy consumption of the intelligent building; Solve the optimal solution of the energy consumption optimization objective function to obtain the minimum energy consumption, and determine the working parameters of the energy consumption equipment according to the minimum energy consumption to control the energy consumption equipment.

9. A non-volatile computer storage medium stores computer-executable instructions, characterized in that, The computer-executable instructions are set as: For the current time node, obtain the historical data of the intelligent building and the historical working parameters of the energy consumption equipment in the previous preset time period; the historical data includes historical indoor data and historical outdoor data; the historical indoor data includes historical indoor environmental data and historical energy consumption data, and the historical outdoor data includes historical outdoor environmental data; Input the historical data and the historical working parameters into each deployed large power industry model to obtain the first predicted indoor data of each large power industry model for the current preset time period; Obtain the actual indoor data of the intelligent building and the current working parameters of the energy-consuming equipment within the current preset time period; According to the first predicted indoor data and the actual indoor data, respectively obtain the prediction errors of each large power industry model, and select multiple preferred large power industry models with prediction errors lower than the preset threshold; Input the actual indoor data and the current working parameters into the preferred large power industry models to obtain the future indoor data for the next preset time period, and calculate the average values corresponding to the future indoor data respectively; Based on the average values, determine the building conditions for the next preset time period, and match the adapted large industry model corresponding to the building conditions in the model library; Based on the average values, determine the building conditions for the next preset time period, specifically including: Determine the indoor environmental conditions of the intelligent building in the next preset time period according to the first average value of the future indoor environmental data; Determine the energy consumption conditions of the intelligent building in the next preset time period according to the second average value of the future energy consumption data; Based on the indoor environmental conditions and the energy consumption conditions, determine the building conditions for the next preset time period; According to the second predicted indoor data output by the adapted large industry model for the next preset time period, with the condition that the indoor environmental value is within the preset comfort range, establish an energy consumption optimization objective function for solving the minimum energy consumption of the intelligent building; Solve the optimal solution of the energy consumption optimization objective function to obtain the minimum energy consumption, and determine the working parameters of the energy-consuming equipment according to the minimum energy consumption to regulate the energy-consuming equipment.

Images