Method, device, electronic device and storage medium for determining building spatiotemporal sequence temperature

Abstract

The present invention discloses a method, device, electronic device, and storage medium for determining the spatiotemporal temperature of a building. The method determines the average temperature time series data corresponding to a floor based on the temperature data sequence collected at each monitoring point; selects a specified number of monitoring points from the monitoring points to form a plurality of monitoring point subsets; and determines a target subset within the plurality of monitoring point subsets based on first correlation data between the temperature data sequences of the monitoring points in the monitoring point subsets and the average temperature time series data, as well as second correlation data between the temperature data sequences of the monitoring points in the monitoring point subsets; and then determines the temperature of the floor based on the temperature data sequence of the representative monitoring points in the target subset, thereby more accurately determining the comprehensive temperature of the building and meeting the different needs of users in the building for heating and cooling comfort.

Description

Technical Field

[0001] The present invention relates to the technical field of building temperature automation control, and in particular to a method, device, electronic equipment and storage medium for determining the spatiotemporal sequence temperature of a building. Background Art

[0002] In building load forecasting-based control technologies, the cooling and heating loads of a building's cooling and heating systems are typically characterized by calculating the building's integrated room temperature. Common methods for calculating the integrated room temperature include the averaging method.

[0003] In related technologies, temperature monitoring points are randomly installed throughout a building, and the average of the monitored temperatures is then calculated. While this method can provide a baseline temperature reading for the entire building, the accuracy of the resulting comprehensive building temperature needs to be improved to meet the varying heating and cooling comfort requirements of building users. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems in the related art to a certain extent. To this end, the present invention provides a method, device, electronic device and storage medium for determining the temporal and spatial sequence temperature of a building.

[0005] The present invention provides a method for determining the spatiotemporal temperature sequence of a building, wherein the building includes at least one floor, each floor corresponds to a temperature monitoring area, the temperature monitoring area is divided into a plurality of monitoring areas, monitoring points are deployed within the monitoring areas, and temperature sensors are installed at the monitoring points; the method comprises:

[0006] Determining the average temperature time series data corresponding to the floor based on the temperature data sequence collected at each of the monitoring points;

[0007] Selecting a specified number of monitoring points from the monitoring points to form a plurality of monitoring point subsets;

[0008] determining a target subset from the plurality of monitoring point subsets based on first correlation data between temperature data sequences of monitoring points in the monitoring point subsets and the average temperature time series data, and second correlation data between temperature data sequences of monitoring points in the monitoring point subsets; wherein the target subset includes representative monitoring points that can be used to characterize the floor temperature;

[0009] The temperature of the floor is determined based on the temperature data sequence of representative monitoring points in the target subset.

[0010] In one embodiment, determining the average temperature time series data corresponding to the floor based on the temperature data sequence collected at each monitoring point includes:

[0011] Determining a valid monitoring point set from the monitoring points based on the initial temperature data sequence collected at each of the monitoring points; wherein the valid monitoring point set includes a plurality of valid monitoring points;

[0012] Based on the valid temperature data sequence collected at each of the valid monitoring points, the average temperature time series data corresponding to the floor is determined.

[0013] In one embodiment, selecting a specified number of monitoring points from the monitoring points to form a plurality of monitoring point subsets includes:

[0014] Taking one third of the number of valid monitoring points in the valid monitoring point set as the designated number, the valid monitoring points included in the valid monitoring point set are arranged and combined to obtain the plurality of monitoring point subsets.

[0015] In one embodiment, determining a target subset from the plurality of monitoring point subsets based on first correlation data between temperature data sequences of monitoring points in the monitoring point subset and the average temperature time series data, and second correlation data between temperature data sequences of monitoring points in the monitoring point subset, includes:

[0016] Determine an average value of first mutual information between the temperature data sequence of each monitoring point in the monitoring point subset and the average temperature time series data as first relevant data of the monitoring point subset;

[0017] determining, based on the temperature data sequences of the monitoring points in the monitoring point subset, an average value of the second mutual information between the monitoring points in the monitoring point subset as the second relevant data of the monitoring point subset;

[0018] According to the average value of the first mutual information and the average value of the second mutual information, a target subset that meets a preset correlation condition is determined from the plurality of monitoring point subsets.

[0019] In one embodiment, determining a target subset that meets a preset correlation condition from among the plurality of monitoring point subsets based on the average value of the first mutual information and the average value of the second mutual information includes:

[0020] For any subset of monitoring points, determining target-related data of the subset of monitoring points based on an average value of the first mutual information of the subset of monitoring points and an average value of the second mutual information of the subset of monitoring points;

[0021] Among the plurality of monitoring point subsets, a monitoring point subset whose target data satisfies a maximum correlation minimum redundancy condition is determined as the target subset.

[0022] In one embodiment, the method further includes: drawing a monitoring point image of the floor based on representative monitoring points in the target subset.

[0023] In one embodiment, the representative monitoring point corresponds to a population density; the method further includes: based on the population density corresponding to the representative monitoring point, displaying a marked object corresponding to the population density at the position of the representative monitoring point in the monitoring point image.

[0024] In one embodiment, determining the temperature of the floor based on the temperature data sequence representing the monitoring points in the target subset includes:

[0025] The temperature data sequence of the representative monitoring points is input into the room temperature prediction neural network model corresponding to the floor for prediction to obtain the comprehensive room temperature of the floor.

[0026] In one embodiment, the monitoring area is divided according to the identification information of the temperature sensor; the floor corresponds to a floor category, and the floor category is any one of the top floor, upper middle floor, lower middle floor, bottom floor, and underground.

[0027] The present invention provides a device for determining the spatiotemporal temperature sequence of a building. The building includes at least one floor, each floor has a corresponding temperature monitoring area, the temperature monitoring area is divided into a plurality of monitoring areas, monitoring points are deployed within the monitoring areas, and temperature sensors are installed at the monitoring points. The device includes:

[0028] an average temperature determination module, configured to determine the average temperature time series data corresponding to the floor based on the temperature data sequence collected at each of the monitoring points;

[0029] A monitoring point subset formation module, configured to select a specified number of monitoring points from the monitoring points to form a plurality of monitoring point subsets;

[0030] a target subset determining module, configured to determine a target subset from the plurality of monitoring point subsets based on first correlation data between temperature data sequences of monitoring points in the monitoring point subsets and the average temperature time series data, and second correlation data between temperature data sequences of monitoring points in the monitoring point subsets; wherein the target subset includes representative monitoring points that can be used to characterize the floor temperature;

[0031] The floor temperature determination module is used to determine the temperature of the floor based on the temperature data sequence representing the monitoring points in the target subset.

[0032] The present invention provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0033] The present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of any one of the above-mentioned methods when executed by a processor.

[0034] In the above embodiment, the average temperature time series data corresponding to the floor is determined based on the temperature data sequence collected at each of the monitoring points; and a specified number of monitoring points are selected from the monitoring points to form a number of monitoring point subsets; thereby, a target subset is determined in the number of monitoring point subsets based on the first correlation data between the temperature data sequence of the monitoring points in the monitoring point subset and the average temperature time series data, and the second correlation data between the temperature data sequences of the monitoring points in the monitoring point subset; and then the temperature of the floor is determined based on the temperature data sequence of the representative monitoring points in the target subset, so as to more accurately determine the comprehensive temperature of the building and meet the different needs of users in the building for hot and cold comfort.

[0035] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1a Schematic diagram of an application scenario of a method for determining the spatiotemporal sequence temperature of a building provided in accordance with an embodiment of this specification.

[0037] Figure 1b The present invention is a flowchart of a method for determining the spatiotemporal sequence temperature of a building provided in accordance with an embodiment of the present specification.

[0038] Figure 1c Schematic diagram of monitoring point images provided according to an embodiment of this specification.

[0039] Figure 2 The present invention is a flowchart of a method for determining the spatiotemporal sequence temperature of a building according to one embodiment of the present specification.

[0040] Figure 3 The present invention is a flowchart of determining the average temperature time series data corresponding to a floor according to one embodiment of the present invention.

[0041] Figure 4 The present invention is a schematic diagram of a process for determining a target subset from a plurality of monitoring point subsets according to one embodiment of the present specification.

[0042] Figure 5The present invention is a flowchart of determining a target subset that meets a preset relevance condition according to one embodiment of the present invention.

[0043] Figure 6 The present invention is a structural block diagram of a device for determining the spatiotemporal temperature of a building according to one embodiment of the present invention. DETAILED DESCRIPTION

[0044] The following describes embodiments of the present invention in detail, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and are not to be construed as limiting the present invention.

[0045] In my country, building operating energy consumption and total process energy consumption account for 21.7% and 46.5% of the nation's total energy consumption, respectively. With the deepening of industrialization and urbanization in my country, total building energy consumption is expected to increase further. Therefore, energy conservation and emission reduction in building operating energy are crucial aspects of this work. Currently, buildings in industrial parks and industrial enterprise buildings (hereinafter referred to as parks) are commonly plagued by energy consumption issues such as extensive energy management, low energy efficiency, high energy consumption, and increased environmental pressure. The public cooling and heating systems in existing older park buildings have independent group control systems, enabling one-touch start / stop and interlocking control of cooling and heating units and water pumps. However, most older park buildings lack water temperature control strategies for their chillers, or suffer from crude chiller load-on / load-off strategies, reliance on manual control, and inexperienced operators. Furthermore, the public cooling and heating systems in older park buildings often lack cooling and heating system energy efficiency monitoring and analysis tools, failing to effectively ensure comfortable and stable room temperatures. Furthermore, they lack energy-efficient operation control strategies, leaving significant room for improvement in energy-saving and optimized operation.

[0046] Typically, cooling and heating systems in older campus buildings use a common, pre-set control logic. This logic calculates the actual building load based on the chilled water supply and return temperature difference and water flow rate, determining whether additional chillers are needed. This approach fails to automatically adjust operating parameters based on factors such as outdoor climate, indoor terminal temperature, sunlight exposure, holidays, and occupancy density. This results in inefficient cooling and heating systems, making them unable to meet the heating and cooling comfort needs of some users and leading to energy waste.

[0047] Therefore, control technologies based on building load forecasting are often used to optimize the energy efficiency of public cooling and heating systems in campus buildings. Building load forecasting technologies can be categorized as physical model-based and algorithmic model-based. Physical model-based building load forecasting technologies typically use heat balance equations and dynamic thermodynamic models such as thermal resistance-capacitance networks to predict the cooling and heating loads of building cooling and heating systems. Algorithmic model-based building load forecasting technologies typically design building load characteristics based on multiple factors, such as the building's outdoor climate and historical cooling and heating supply. Machine learning models are then used to establish a mapping between these building load characteristics and cooling and heating loads. However, due to the complex structures of buildings and the significant variations in physical models, physical model-based building load forecasting methods have limited scalability and generalizability. With the rapid development of machine learning and the widespread integration of a large number of IoT sensors in buildings, algorithmic model-based building load forecasting technologies have achieved significant development and application. However, current technologies lack the ability to extract features from factors such as indoor terminal temperature, outdoor weather, holidays, and population density. They are unable to adapt to the randomness and variability of user behavior in buildings and cannot meet users' cooling and heating comfort needs, thus leading to energy waste.

[0048] In related technologies, in control technologies based on building load prediction, the cooling and heating loads of a building's cooling and heating source systems are usually characterized by calculating the building's comprehensive room temperature. There are two common methods for calculating the comprehensive room temperature:

[0049] (1) Averaging method. This method randomly installs temperature monitoring points and then averages the monitored temperatures. This method can reflect the basic temperature level of the entire building, but it cannot perform differentiated analysis and cannot meet the cooling and heating comfort needs of some users, such as basement property staff, rooftop users, and north-facing users, which can easily lead to user complaints.

[0050] (2) Weight method. This method divides the entire building into three categories of users: the first category is the top floor, the second category is the middle household, and the third category is the side household and the ground floor household. By installing typical monitoring points for the three categories of users, for example, at least one room temperature monitoring point is installed for each of the first category of users; at least two types of room temperature monitoring points are installed for each of the second category of users, one type of monitoring point provides heating to the users upstairs, and the other type of monitoring point provides no heating to the users upstairs; at least two types of room temperature monitoring points are installed for each of the third category of users, one type of monitoring point provides heating to the users upstairs, and the other type of monitoring point provides no heating to the users upstairs. Thus, at least five types of room temperature monitoring points are installed in total. Then, a weight is set for each type of user, and finally, the comprehensive room temperature of the entire building is calculated based on the weight. This method can better reflect the heating efficiency of the entire building, but the division of typical users is too rough and cannot accurately reflect the actual heating situation of the building. At the same time, this method only targets the building heating system and does not truly reflect the building cooling demand (for example, the cooling and heating demands of users on the sunny side are opposite). Therefore, it cannot well meet the cooling and heating comfort needs of users in the park buildings.

[0051] To adapt the comprehensive room temperature calculation method for buildings (hereinafter referred to as "buildings") to the complex and volatile building user structure, it is necessary to provide a method, device, electronic device, and storage medium for determining the spatiotemporal temperature of a building. This method topologically partitions the building's floors and the locations of the monitoring points contained within each floor. Representative monitoring points are then selected from multiple temperature sampling monitoring points on each branch floor (each floor of a secondary topological branch). These representative monitoring points are then used to characterize the indoor terminal temperature, outdoor weather, holidays, and occupancy density of each branch floor. The representative monitoring points are then used to calculate the comprehensive room temperature of each building floor in real time. This is then used to determine optimal operational decisions for the building's cooling and heating systems through load forecasting, data analysis, and operational strategy optimization, thereby achieving energy conservation and consumption reduction. This method can meet the cooling and heating comfort requirements of rooms on each branch floor without requiring individual modeling of each room within the building. This reduces the number of indoor temperature monitoring devices used, thereby reducing algorithmic errors caused by data transmission errors from a large number of devices. It also conserves computational and data storage resources for the comprehensive room temperature calculation method.

[0052] Figure 1a This is a schematic diagram of an application scenario for a method for determining a building's spatiotemporal temperature sequence, as provided in this specification. Taking Building A, a 15-story campus building, as an example, a temperature sensor can be installed in each room on each floor to collect temperature data sequences for each room. It should be noted that each temperature sensor can be assigned a unique number, which is used to identify the location of its monitoring point. Figure 1aThis is a schematic diagram of the topological structure obtained by topologically dividing each floor in Building A of the campus and the monitoring point locations of the temperature sensors on each floor.

[0053] In this scenario example, first, the floors of Building A in the campus are divided into a first-level topology structure of top floor, upper-middle floor, lower-middle floor, bottom floor, and underground; secondly, the monitoring point locations of the temperature sensors contained in each floor are divided into a second-level topology according to the structure of each floor; again, representative monitoring points corresponding to each floor are selected from the monitoring point locations of the temperature sensors contained in each floor; then, based on the temperature data sequence collected by the representative monitoring points corresponding to each floor and the average temperature time series data corresponding to each floor, the room temperature prediction neural network model corresponding to each floor is trained; finally, based on the room temperature prediction neural network model corresponding to each floor, the comprehensive room temperature corresponding to each floor can be calculated.

[0054] The following example illustrates how to divide the first-level topology. In this scenario example, based on the lighting conditions of each floor (including lighting time, lighting intensity, etc.), each floor of Building A is divided into a first-level topology including a top-level branch (sufficient lighting, long lighting time), an upper-middle-level branch (relatively sufficient lighting, long lighting time), a lower-middle-level branch (average lighting, average lighting time), a bottom-level branch (less lighting, short lighting time), and an underground branch (no lighting). In some embodiments, if the top floor has the same structure as other floors, it is not necessary to divide the top floor separately.

[0055] This example illustrates how to divide the secondary topology. In this scenario, the top branch includes the 15th floor, the upper-middle branch includes floors 9-14, the lower-middle branch includes floors 3-8, the bottom branch includes floors 1-2, and the underground branch includes buildings B1 and B2. Based on the structure of each floor, the temperature sensors corresponding to each floor in the primary topology branch are divided into secondary topologies.

[0056] In this scenario, each floor of Building A on the campus can include a north-facing location, a south-facing location, and a center location. For the top-level branch, because the center of the 15th floor has a staircase leading to the rooftop, it will affect the temperature data of the rooms in the center of the floor and require additional attention. Therefore, the monitoring point locations of the temperature sensors on the 15th floor of the top-level branch are divided into two levels: north-facing, south-facing, and center-facing. For the upper-middle and lower-middle branches, because the temperature data at the center of each floor is similar to the average temperature time series data of each floor, no additional attention is required. Therefore, the monitoring point locations of the temperature sensors on each floor in the upper-middle and lower-middle branches are divided into two levels: north-facing and south-facing. For the bottom branch, since floors 1-2 mainly include the basement and lobby, and the temperature data of the basement is usually affected by the density of people, the cooling and heating comfort requirements of the north and south sides are different. The temperature data of the lobby is usually close to the average temperature time series data of the floor. Therefore, the monitoring point locations of the temperature sensors contained in the bottom branch are divided into a secondary topology according to the lobby and the basement. For the underground B1-B2 floors, since the cooling and heating comfort requirements of these two floors are related to the depth of the floor, the monitoring point locations of the temperature sensors contained in the underground branch are directly divided into a secondary topology according to the floor. Please continue to refer to Figure 1a ,It should be noted that the number of monitoring point locations contained on each floor in the brackets “()” is the same as the number of ,monitoring point locations contained on the top floor.

[0057] In this scenario example, according to the population density in the rooms corresponding to the monitoring point locations, the monitoring point locations corresponding to rooms with lower population density can be represented by squares, and the monitoring point locations corresponding to other rooms can be represented by circles.

[0058] The following example illustrates how to select a representative monitoring point. In this scenario, first, based on the temperature data collected by the temperature sensors on each floor, the average temperature time series data for each floor is calculated. Second, for any secondary topology branch, a certain number of arbitrary monitoring point locations are selected from the monitoring point locations contained in the secondary topology branch to form several monitoring point subsets. Third, for any monitoring point subset in the several monitoring point subsets, the first correlation data between the temperature data of each monitoring point location contained in the arbitrary monitoring point subset and the average temperature time series data of the floor on which the monitoring point location is located is calculated, and the average value of all the first correlation data in the arbitrary monitoring point subset is calculated. Third, the second correlation data between the temperature data of each monitoring point location in the arbitrary monitoring point subset is calculated, and the average value of all the second correlation data in the arbitrary monitoring point subset is calculated. Then, the difference between the average value of all the first correlation data and the average value of all the second correlation data corresponding to the arbitrary monitoring point subset is calculated. Finally, based on the above-mentioned differences corresponding to all monitoring point subsets in the secondary topology branch, the monitoring point location in the monitoring point subset corresponding to the largest difference is determined as the representative monitoring point in the secondary topology branch.

[0059] In this scenario, actual analysis revealed that the average temperature data for each floor in the upper-middle branch is similar. Therefore, for the upper-middle branch, the average temperature time series data for only one floor can be calculated to select representative monitoring points for multiple floors in the upper-middle branch. The representative monitoring point selection method for the lower-middle branch is the same as for the upper-middle branch.

[0060] See also Figure 1b In this scenario, before selecting representative monitoring points, valid monitoring points can be selected. Specifically, for all collected temperature data, default and unreasonable values ​​are first eliminated. Secondly, the percentage of valid data at each monitoring point relative to the total raw data can be calculated. Monitoring points with valid data accounting for at least 80% of the raw data are selected as valid monitoring points. This prevents abnormal calculations of the comprehensive room temperature on each floor due to distorted temperature data, which in turn fails to effectively reflect the building's true indoor temperature.

[0061] Please continue reading Figure 1b In this scenario, the first correlation data and the second correlation data are calculated using the mutual information theory. A representative monitoring point is selected from the monitoring points of each branch floor (the floor of each secondary topological branch), that is, in the feature set F consisting of m monitoring points m Select a subset S consisting of n (n≤m) monitoring points n According to the maximum correlation principle, the optimal subset S n The monitoring points P included should be satisfied FThe average mutual information of the target variable and the average temperature p of each floor reaches the maximum value. The optimal subset selected only according to the maximum correlation principle has a large redundancy (the various monitoring points in the subset have a large correlation). To this, the constraint condition of the minimum redundancy principle is added, so that S n The average mutual information between the included monitoring points is the smallest. When selecting the representative monitoring points of each branch floor, first calculate the temperature P of each monitoring point F and the mutual information I(P F ,p), and then use the mRMR principle to select representative monitoring points. All combinations and Select a combination that meets the mRMR principle to obtain the optimal subset S n , that is, each branch floor represents a combination of monitoring points.

[0062] Please continue reading Figure 1b This example illustrates how to calculate the comprehensive room temperature for each floor. In this scenario, a neural network model for predicting room temperature for each floor is trained based on historical temperature data from a representative monitoring point on each floor, along with the historical average temperature time series data for that floor. By inputting the temperature data from each floor's representative monitoring point into the trained model, the comprehensive room temperature for each floor is output.

[0063] It should be noted that, in this scenario example, the room temperature prediction neural network model corresponding to each floor in the upper-middle branch can be the same, and the room temperature prediction neural network model corresponding to each floor in the lower-middle branch can be the same.

[0064] Furthermore, in this scenario example, after the representative monitoring points are selected, the monitoring point images of each floor of Building A in the park can be redrawn according to the representative monitoring points of each floor. Figure 1c This is an image of the monitoring points on each floor of Building A, a campus building, drawn based on the representative monitoring points selected for each floor. Based on this image, the rationality of the monitoring point selection can be analyzed and appropriate adjustments can be made.

[0065] It should be noted that in this scenario example, when the population density at any monitoring point on any floor of Building A in the campus changes, or the external environmental conditions of the building change, the monitoring point locations of each floor and the temperature sensors contained on each floor can be topologically re-divided, and new representative monitoring points can be selected.

[0066] The embodiment of this specification provides a method for determining the temporal and spatial sequence temperature of a building. The building includes at least one floor, each floor has a temperature monitoring area, and the temperature monitoring area is divided into several monitoring areas. Monitoring points are deployed in the monitoring areas, and temperature sensors are installed at the monitoring points. Figure 2 The method for determining the temperature of a building's spatiotemporal sequence may include the following steps:

[0067] S210. Determine the average temperature time series data corresponding to each floor based on the temperature data sequence collected at each monitoring point.

[0068] The building can be an office building or industrial enterprise building, and each floor has a corresponding temperature monitoring area. The temperature monitoring area can be the entire floor or a specific area within the floor. The temperature monitoring area is divided into several monitoring areas. The monitoring areas can be divided based on the office areas of users on the floor or based on the temperature requirements of users on the floor. Monitoring points are deployed within the monitoring area, and temperature sensors are installed at the monitoring points. The temperature sensors collect temperature data within the monitoring area to form a temperature data sequence.

[0069] Specifically, each monitoring point is equipped with a temperature sensor. The temperature data collected by the temperature sensor can be transmitted to a temperature control computer via wireless communication. The computer can directly use the temperature data sequence collected at each monitoring point to calculate the temperature mean, thereby obtaining the average temperature time series data corresponding to each floor. The computer can also preprocess the received temperature data sequence collected at each monitoring point and use the preprocessed temperature data sequence to determine the average temperature time series data corresponding to each floor.

[0070] S220: Select a specified number of monitoring points from the monitoring points to form a plurality of monitoring point subsets.

[0071] The specified number can be determined by the actual number of deployed monitoring points or by the user's actual temperature control requirements. Specifically, a specified number of monitoring points are randomly selected from the plurality of deployed monitoring points, and these specified number of monitoring points constitute a monitoring point subset. Similarly, multiple monitoring point subsets can be obtained. It is understood that some of the monitoring points in a monitoring point subset may overlap.

[0072] S230. Determine a target subset from the plurality of monitoring point subsets based on first correlation data between the temperature data sequences of the monitoring points in the monitoring point subset and the average temperature time series data, and second correlation data between the temperature data sequences of the monitoring points in the monitoring point subset.

[0073] The target subset includes representative monitoring points that can be used to characterize floor temperature.

[0074] In some cases, in order to determine the floor temperature more accurately, it is necessary to select a representative monitoring point from several deployed monitoring points. Therefore, a target subset is selected from several monitoring point subsets based on the correlation data between the temperature data series at the monitoring points to determine the representative monitoring point.

[0075] Specifically, for each monitoring point subset, based on the temperature data sequence of the monitoring points in the monitoring point subset and the average temperature time series data, the first correlation data between the two is calculated. The temperature data sequence of every two monitoring points in the monitoring point subset is used to perform correlation calculation to obtain the second correlation data between the temperature data sequences of the monitoring points in the monitoring point subset. Furthermore, the first correlation data and the second correlation data are used to filter the constructed monitoring point subset to obtain a monitoring point subset that meets the correlation condition requirements, i.e., the target subset. The temperature data collected by the monitoring points in the target subset can be used to characterize the floor temperature, i.e., the monitoring points included in the target subset are representative monitoring points.

[0076] S240 , determining the temperature of the floor based on the temperature data sequence of the representative monitoring points in the target subset.

[0077] Specifically, after determining the representative monitoring points, the temperature of each floor can be determined using the temperature data sequence of the representative monitoring points in the target subset. This temperature data sequence can be input into a neural network model for prediction, and the temperature output by the neural network model is determined as the floor temperature. Furthermore, the comprehensive floor room temperature calculated from the representative monitoring points can be used through load forecasting, data analysis, and operational strategy optimization to develop appropriate operational decision plans for the cooling and heating source systems on each floor. This effectively ensures comfortable indoor temperatures for users, achieving energy conservation and consumption reduction for the entire building, energy conservation and emission reduction for the park, and environmental protection.

[0078] In the above embodiment, the average temperature time series data corresponding to each floor is determined based on the temperature data sequence collected at each monitoring point. A specified number of monitoring points are selected from the monitoring points to form several monitoring point subsets. A target subset is then determined from the several monitoring point subsets based on the first correlation data between the temperature data sequences of the monitoring points in the monitoring point subsets and the average temperature time series data, as well as the second correlation data between the temperature data sequences of the monitoring points in the monitoring point subsets. Furthermore, the temperature of each floor is determined based on the temperature data sequence of the representative monitoring points in the target subset, more accurately determining the overall building temperature and meeting the varying heating and cooling comfort requirements of building users. Furthermore, while effectively ensuring comfortable indoor temperatures for users, energy conservation and consumption reduction are achieved throughout the building, achieving energy conservation, emission reduction, and environmental protection.

[0079] In some embodiments, see Figure 3 , based on the temperature data sequence collected at each monitoring point, determining the average temperature time series data corresponding to the floor may include the following steps:

[0080] S310 : Determine a valid monitoring point set among the monitoring points based on the initial temperature data sequence collected at each monitoring point.

[0081] S320. Determine the average temperature time series data corresponding to each floor based on the valid temperature data sequence collected at each valid monitoring point.

[0082] Among them, the effective monitoring point set includes several effective monitoring points. Specifically, movable temperature and humidity sensors (4G) are placed as comprehensively as possible on each floor to collect the user's indoor temperature data for multiple consecutive days to obtain an initial temperature data sequence. Determine whether there are default values ​​and unreasonable values ​​in the initial temperature data sequence. After eliminating the abnormal data in the above situations, calculate the percentage of effective data in the total amount of original data, screen out data monitoring points where the effective data accounts for more than 80% of the original data, and determine the effective monitoring point set among several monitoring points. The monitoring points included in the effective monitoring point set are effective monitoring points. Eliminate monitoring points with weak temperature sensor signals or abnormal data to prevent abnormal comprehensive room temperature calculation due to temperature data distortion, which cannot effectively reflect the true situation of indoor temperature. After eliminating abnormal data, use the effective temperature data sequence collected at each effective monitoring point to calculate the average temperature time series data corresponding to the floor.

[0083] In the above embodiment, by determining a set of valid monitoring points and using the valid temperature data sequence collected at the valid monitoring points in the valid monitoring point set, the average temperature time series data corresponding to the floor is determined, thereby reducing the impact of abnormal data on the calculation of the comprehensive room temperature of the floor and improving the accuracy of the determined comprehensive room temperature of the floor.

[0084] In some embodiments, selecting a specified number of monitoring points from the monitoring points to form several monitoring point subsets may include: taking one third of the number of valid monitoring points in the valid monitoring point set as the specified number, and arranging and combining the valid monitoring points included in the valid monitoring point set to obtain several monitoring point subsets.

[0085] Specifically, the number of valid monitoring points in the valid monitoring point set is denoted as m. N valid monitoring points are obtained from the valid monitoring point set and permuted and combined to obtain a number of monitoring point subsets. Here, n is equal to one-third of m. For example, if the number of valid monitoring points is 6, any two of the 6 valid monitoring points are used to form a monitoring point subset.

[0086] In some embodiments, see Figure 4, determining a target subset from a plurality of monitoring point subsets based on first correlation data between temperature data sequences of monitoring points in the monitoring point subset and average temperature time series data, and second correlation data between temperature data sequences of monitoring points in the monitoring point subset, may include:

[0087] S410 , determining an average value of first mutual information between the temperature data sequence of each monitoring point in the monitoring point subset and the average temperature time series data as first relevant data of the monitoring point subset.

[0088] S420 : Based on the temperature data sequences of the monitoring points in the monitoring point subset, determine an average value of the second mutual information between the monitoring points in the monitoring point subset as the second relevant data of the monitoring point subset.

[0089] S430: Determine a target subset that meets a preset correlation condition from a plurality of monitoring point subsets according to an average value of the first mutual information and an average value of the second mutual information.

[0090] Specifically, the mutual information (MI) theory is used to select the target subset from several monitoring point subsets. In each floor, based on the temperature data sequence N of each monitoring point in a certain time period, fp And the corresponding floor average temperature time series data N for this period rp , calculate the mutual information between the temperature data series of each monitoring point and the average floor temperature time series data. The calculation formula is as follows:

[0091]

[0092] Among them, N is the total number of data of the average temperature series of monitoring points on each floor. In addition, when N fp When (i,j)=0, it is not counted. F Represents the monitoring point on the p floor.

[0093] The average value of the first mutual information between the temperature data sequence of each monitoring point in the monitoring point subset and the average temperature time series data is calculated as the first correlation data for the monitoring point subset. The average value of the second mutual information between the monitoring points in the monitoring point subset is calculated using the temperature data sequences of the monitoring points in the monitoring point subset, and this is used as the second correlation data for the monitoring point subset. The monitoring point subset is screened based on the average values ​​of the first mutual information and the second mutual information to obtain a target subset that meets the preset correlation conditions.

[0094] In some embodiments, see Figure 5 , based on the average value of the first mutual information and the average value of the second mutual information, determining a target subset that meets a preset correlation condition from a plurality of monitoring point subsets may include:

[0095] S510 : For any subset of monitoring points, determine target-related data of any subset of monitoring points based on an average value of first mutual information of any subset of monitoring points and an average value of second mutual information of any subset of monitoring points.

[0096] S520 . Determine, from among the plurality of monitoring point subsets, a monitoring point subset whose target data satisfies the maximum correlation and minimum redundancy conditions as a target subset.

[0097] Among them, the mutual information (MI) theory is adopted to select the optimal feature subset based on the maximum correlation-minimum redundancy (mRMR) feature method. In this way, the selected monitoring points on each floor can ensure that the redundant information and noise contained are minimized while maximizing the use of the favorable information of the original data set. The monitoring points contained in the selected feature subset are the representative monitoring points.

[0098] Specifically, in the feature set F composed of m monitoring points m Select a subset S consisting of n (n≤m) monitoring points n According to the maximum correlation principle, the optimal subset S n The monitoring points P included should be satisfied F The average value of the mutual information with the target variable average temperature p of each floor reaches the maximum value, that is,

[0099] The optimal subset selected only according to the maximum correlation principle has a large redundancy (the monitoring points in the subset have a large correlation), and the constraint condition of the minimum redundancy principle is added to make S n The average mutual information between the included monitoring points is the smallest, that is,

[0100] Therefore, first, calculate the temperature P of each monitoring point F and the mutual information I(P F ,p), and then the mRMR principle is used to select representative monitoring points. Specifically, calculate All permutations and combinations and Select a combination that meets the mRMR principle to obtain the target subset S n , that is, each floor represents a combination of monitoring points. The mRMR principle is:

[0101] In the above implementation, the optimal feature subset is selected from multiple sampling monitoring points on each floor of the building by adopting the maximum correlation-minimum redundancy feature selection method. These features can ensure that while maximizing the use of the favorable information of the original data set, the redundant information and noise contained therein are minimized, and representative monitoring points of the comprehensive room temperature on each floor of the campus buildings are effectively found.

[0102] In some embodiments, the method may further include: drawing an image of the monitoring points on the floor based on representative monitoring points in the target subset.

[0103] Specifically, a user monitoring point image can be drawn for each floor. The floor monitoring point image can be drawn in combination with the representative monitoring points in the target subset. Furthermore, the rationality of the selection of representative monitoring points can be analyzed and appropriate adjustments can be made.

[0104] In some embodiments, the representative monitoring point corresponds to a population density. The method may further include: based on the population density corresponding to the representative monitoring point, displaying a marked object corresponding to the population density at a position of the representative monitoring point in the monitoring point image.

[0105] Different occupancy densities can also be displayed within the monitoring point image. Therefore, based on the occupancy density corresponding to the monitoring point, a marker object corresponding to the occupancy density is displayed at the location of the monitoring point. Specifically, marker objects can be represented using different colors or shapes. For example, a room with a low occupancy density can be square, while a room with a high occupancy density can be circular.

[0106] In some embodiments, determining the temperature of a floor based on a temperature data sequence representing a monitoring point in a target subset may include: inputting the temperature data sequence representing the monitoring point into a room temperature prediction neural network model corresponding to the floor for prediction to obtain a comprehensive room temperature of the floor.

[0107] Different floors can correspond to different room temperature prediction neural network models. Specifically, for different floors, the temperature data sequence representing the monitoring points on that floor is input into the room temperature prediction neural network model corresponding to that floor for prediction to obtain the comprehensive room temperature of that floor.

[0108] For example, after selecting representative monitoring points on each floor, a comprehensive room temperature model for each floor is established, and a neural network (ANN) model is used. During training, the temperature P of the representative monitoring point on each floor is input. F And the average temperature p of each floor is used for learning and training, and the error indicator is the root mean square error:

[0109]

[0110] Where: P Fiis the temperature of the representative monitoring point at time i, p i is the average temperature of the floor at time i, and N is the number of data points. When calculating the comprehensive room temperature of a floor, the temperature of the representative monitoring point on that floor is substituted into the trained model to output the comprehensive room temperature of that floor.

[0111] In some embodiments, the temperature P of each floor representative monitoring point can also be used F The optimal weight coefficients for each floor's representative monitoring points are obtained through training using a cuckoo search algorithm, along with the average temperature p of each floor. When calculating the comprehensive room temperature on each floor, the comprehensive room temperature on each floor can be obtained based on the temperatures of the representative monitoring points on each floor and the optimal weight coefficients corresponding to these points.

[0112] In some other embodiments, a weight coefficient for each floor can be calculated based on the comprehensive room temperature of each floor and the overall average temperature of the building. Based on the comprehensive room temperature of each floor and the weight coefficient corresponding to each floor, the overall comprehensive room temperature of the building can be obtained.

[0113] In the above implementation, the comprehensive room temperature of each floor is obtained by statistical calculation of representative monitoring points, and the best operation decision plan for the cold and heat source systems on each floor can be given through load forecasting, data analysis and operation strategy optimization, so as to effectively ensure the comfortable indoor temperature of users and achieve the goals of energy saving and consumption reduction for the entire building, energy saving and emission reduction for the park and environmental protection.

[0114] In some embodiments, the monitoring area is divided according to the identification information of the temperature sensor. Each floor has a floor category, which is any one of the top floor, upper middle floor, lower middle floor, bottom floor, and underground.

[0115] Specifically, the monitoring areas of each floor can be divided according to floor categories, and the selected representative monitoring points can be the representative monitoring points in each floor category. Furthermore, the monitoring areas in each floor category can be divided according to characteristic categories such as north-facing, south-facing, and centered, and the selected representative monitoring points can be the representative monitoring points in each characteristic category.

[0116] The embodiment of this specification provides a device for determining the temporal and spatial sequence temperature of a building. The building includes at least one floor, each floor has a corresponding temperature monitoring area, and the temperature monitoring area is divided into several monitoring areas. Monitoring points are deployed in the monitoring areas, and temperature sensors are installed at the monitoring points. Figure 6 The building spatiotemporal sequence temperature determination device includes: an average temperature determination module, a monitoring point subset composition module, a target subset determination module, and a floor temperature determination module.

[0117] an average temperature determination module, configured to determine the average temperature time series data corresponding to the floor based on the temperature data sequence collected at each of the monitoring points;

[0118] A monitoring point subset formation module, configured to select a specified number of monitoring points from the monitoring points to form a plurality of monitoring point subsets;

[0119] a target subset determining module, configured to determine a target subset from the plurality of monitoring point subsets based on first correlation data between temperature data sequences of monitoring points in the monitoring point subsets and the average temperature time series data, and second correlation data between temperature data sequences of monitoring points in the monitoring point subsets; wherein the target subset includes representative monitoring points that can be used to characterize the floor temperature;

[0120] The floor temperature determination module is used to determine the temperature of the floor based on the temperature data sequence representing the monitoring points in the target subset.

[0121] The specific definitions of the apparatus for determining the temporal and spatial sequence temperature of a building can be found in the definitions of the method for determining the temporal and spatial sequence temperature of a building, as described above, and will not be further elaborated here. Each module within the apparatus for determining the temporal and spatial sequence temperature of a building can be implemented in whole or in part via software, hardware, or a combination thereof. Each of these modules can be embedded in or independent of a processor within a computer device in hardware form, or stored in a computer device memory in software form, allowing the processor to call and execute the corresponding operations of each module.

[0122] The embodiments of this specification also provide an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for determining the spatiotemporal temperature sequence of a building according to any of the aforementioned embodiments is implemented.

[0123] The embodiments of this specification also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for determining the spatiotemporal temperature sequence of a building according to any of the aforementioned embodiments.

[0124] It should be noted that the logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic device), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and portable compact disc read-only memory (CDROM). Furthermore, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing it in another suitable manner if necessary, and then storing it in a computer memory.

[0125] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0126] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0127] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0128] In the present invention, unless otherwise specified or limited, the terms "installed," "connected," "connect," "fixed," etc. should be understood in a broad sense. For example, they can refer to fixed connection, detachable connection, or integration; mechanical connection, electrical connection; direct connection, or indirect connection through an intermediate medium; internal communication between two components, or interaction between two components, unless otherwise specified. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0129] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A method for determining the temperature of a building in a spatiotemporal sequence, characterized in that: The building includes at least one floor, each floor corresponds to a temperature monitoring area, the temperature monitoring area is divided into a plurality of monitoring areas, monitoring points are deployed in the monitoring areas, and temperature sensors are installed at the monitoring points; the method includes: Determining the average temperature time series data corresponding to the floor based on the temperature data sequence collected at each of the monitoring points; Selecting a specified number of monitoring points from the monitoring points to form a plurality of monitoring point subsets; determining a target subset from the plurality of monitoring point subsets based on first correlation data between temperature data sequences of monitoring points in the monitoring point subsets and the average temperature time series data, and second correlation data between temperature data sequences of monitoring points in the monitoring point subsets; wherein the target subset includes representative monitoring points that can be used to characterize the floor temperature; determining the temperature of the floor based on a temperature data sequence representing a monitoring point in the target subset; Determining a target subset from the plurality of monitoring point subsets based on first correlation data between the temperature data sequences of the monitoring points in the monitoring point subsets and the average temperature time series data, and second correlation data between the temperature data sequences of the monitoring points in the monitoring point subsets, comprises: Determine an average value of first mutual information between the temperature data sequence of each monitoring point in the monitoring point subset and the average temperature time series data as first relevant data of the monitoring point subset; determining, based on the temperature data sequences of the monitoring points in the monitoring point subset, an average value of the second mutual information between the monitoring points in the monitoring point subset as the second relevant data of the monitoring point subset; Determining a target subset that meets a preset correlation condition from the plurality of monitoring point subsets according to an average value of the first mutual information and an average value of the second mutual information; The determining, based on the average value of the first mutual information and the average value of the second mutual information, a target subset that meets a preset correlation condition from the plurality of monitoring point subsets includes: For any subset of monitoring points, determining target-related data of the subset of monitoring points based on an average value of the first mutual information of the subset of monitoring points and an average value of the second mutual information of the subset of monitoring points; Among the plurality of monitoring point subsets, a monitoring point subset whose target data satisfies a maximum correlation minimum redundancy condition is determined as the target subset.

2. The method according to claim 1, characterized in that The step of determining the average temperature time series data corresponding to the floor based on the temperature data sequence collected at each monitoring point includes: Determining a valid monitoring point set from the monitoring points based on the initial temperature data sequence collected at each of the monitoring points; wherein the valid monitoring point set includes a plurality of valid monitoring points; Based on the valid temperature data sequence collected at each of the valid monitoring points, the average temperature time series data corresponding to the floor is determined.

3. The method according to claim 2, characterized in that The selecting a specified number of monitoring points from the monitoring points to form a plurality of monitoring point subsets includes: Taking one third of the number of valid monitoring points in the valid monitoring point set as the designated number, the valid monitoring points included in the valid monitoring point set are arranged and combined to obtain the plurality of monitoring point subsets.

4. The method according to claim 1, wherein The method further comprises: Based on the representative monitoring points in the target subset, an image of the monitoring points of the floor is drawn.

5. The method according to claim 4, characterized in that The representative monitoring point corresponds to a population density; the method further includes: Based on the population density corresponding to the representative monitoring point, a marked object corresponding to the population density is displayed at the position of the representative monitoring point in the monitoring point image.

6. The method according to claim 1, wherein Determining the temperature of the floor based on the temperature data sequence representing the monitoring points in the target subset includes: The temperature data sequence of the representative monitoring points is input into the room temperature prediction neural network model corresponding to the floor for prediction to obtain the comprehensive room temperature of the floor.

7. The method according to any one of claims 1 to 6, characterized in that The monitoring area is divided according to the identification information of the temperature sensor; the floors correspond to floor categories, and the floor categories are any one of the top floor, upper middle floor, lower middle floor, bottom floor, and underground.

8. A device for determining the temporal and spatial sequence temperature of a building, characterized in that: The building includes at least one floor, each floor has a corresponding temperature monitoring area, the temperature monitoring area is divided into a plurality of monitoring areas, monitoring points are deployed in the monitoring areas, and temperature sensors are installed at the monitoring points; the device includes: an average temperature determination module, configured to determine the average temperature time series data corresponding to the floor based on the temperature data sequence collected at each of the monitoring points; A monitoring point subset formation module, configured to select a specified number of monitoring points from the monitoring points to form a plurality of monitoring point subsets; a target subset determining module, configured to determine a target subset from the plurality of monitoring point subsets based on first correlation data between temperature data sequences of monitoring points in the monitoring point subsets and the average temperature time series data, and second correlation data between temperature data sequences of monitoring points in the monitoring point subsets; wherein the target subset includes representative monitoring points that can be used to characterize the floor temperature; A floor temperature determination module, configured to determine the temperature of the floor based on a temperature data sequence representing a monitoring point in the target subset; The target subset determination module determines a target subset in the plurality of monitoring point subsets based on first correlation data between the temperature data sequences of the monitoring points in the monitoring point subset and the average temperature time series data, and second correlation data between the temperature data sequences of the monitoring points in the monitoring point subset, specifically comprising: determining an average value of first mutual information between a temperature data sequence of each monitoring point in the monitoring point subset and the average temperature time series data as first relevant data for the monitoring point subset; determining an average value of second mutual information between the monitoring points in the monitoring point subset based on the temperature data sequences of the monitoring points in the monitoring point subset as second relevant data for the monitoring point subset; and determining a target subset that meets a preset correlation condition from the plurality of monitoring point subsets based on the average values ​​of the first mutual information and the second mutual information; The target subset determination module determines, based on the average value of the first mutual information and the average value of the second mutual information, a target subset that meets a preset correlation condition from the plurality of monitoring point subsets, specifically including: For any monitoring point subset, the target-related data of any monitoring point subset is determined based on the average value of the first mutual information of any monitoring point subset and the average value of the second mutual information of any monitoring point subset; among the several monitoring point subsets, the monitoring point subset whose target data meets the maximum correlation and minimum redundancy conditions is determined as the target subset.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images