Energy-saving and environment-friendly detection method and system for green building

Abstract

The invention relates to the technical field of energy data processing, in particular to an energy-saving and environment-friendly detection method and system for a green building. According to the change trend of all data of a corresponding dimension in a neighborhood range of data of each dimension at a target sampling moment, the abnormal degree of the data of each dimension at the target sampling moment is obtained; obtaining the connectivity between the two dimension data at the target sampling moment by combining the difference of the abnormal degree between the two dimension data and the local distribution association; further obtaining a correction score coefficient of the power efficiency data at the target sampling moment; correcting the initial score value of the corresponding electrical efficiency data to obtain a corrected score value of the electrical efficiency data at the target sampling moment; obtaining a correction score value of the power efficiency data at each sampling moment; and abnormal electric efficiency data are screened out. According to the method, the accurate score value of each electrical efficiency data is obtained, the accuracy of abnormal electrical efficiency data detection is improved, and the energy-saving and environment-friendly detection result is optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of energy data processing, and in particular to an energy-saving and environmentally friendly detection method and system for green buildings. Background Art

[0002] The energy-saving and environmental protection detection methods of green buildings can improve the energy utilization efficiency of green buildings, reduce carbon emissions and improve the indoor environmental quality. For building energy utilization data, electrical efficiency data is an important component. Electrical efficiency data is electrical data used to characterize the efficiency and performance of buildings in terms of electricity use, and is used to reflect the efficiency and performance of buildings in terms of electricity use; monitoring and analyzing electrical efficiency data can understand the specific situation of the building in terms of electricity consumption, and provide guidance information for building maintenance managers to troubleshoot, thereby effectively reducing energy waste and improving energy conservation and environmental protection levels; therefore, monitoring and analyzing electrical efficiency data in green buildings can reflect the energy waste problem inside the building.

[0003] In the prior art, the Z-score method is used to detect anomalies in the electrical efficiency data in buildings. However, because the electrical efficiency data may be affected by environmental factors such as temperature, electromagnetic interference or improper management to generate noise data, it is necessary to distinguish the noise data from the actual abnormal data. The existing Z-score method uses the same calculation method for the score values ​​of all electrical efficiency data, does not take into account the changes in the data, and cannot accurately distinguish between noise data and abnormal data, resulting in inaccurate abnormal data detection and poor accuracy in energy-saving and environmental protection detection. Summary of the invention

[0004] In order to solve the technical problem that the accuracy of abnormal data detection is poor due to the failure to obtain accurate score values ​​of electrical efficiency data in the prior art, the purpose of the present invention is to provide a method and system for energy-saving and environmental protection detection of green buildings. The technical solutions adopted are as follows: The present invention proposes a green building energy-saving and environmental protection detection method, the method comprising: Acquire multi-dimensional data of the green building at each sampling moment in the sampling time period; the multi-dimensional data includes electrical efficiency data and power factor data; Select any sampling time as the target sampling time; obtain the abnormality of each dimensional data at the target sampling time according to the change trend of all data of the corresponding dimension within the neighborhood range of each dimensional data at the target sampling time; obtain the connection between the two dimensional data at the target sampling time according to the difference in the abnormality between the two dimensional data at the target sampling time and the local distribution association between the two dimensional data; obtain the correction score coefficient of the electric efficiency data at the target sampling time according to the abnormality of the electric efficiency data at the target sampling time and the connection between the two dimensional data; Correcting the initial score value of the corresponding electric efficiency data according to the correction score coefficient of the electric efficiency data at the target sampling time to obtain the corrected score value of the electric efficiency data at the target sampling time; changing the target sampling time to obtain the corrected score value of the electric efficiency data at each sampling time; Abnormal electrical efficiency data is screened out according to the correction score value.

[0005] Furthermore, the method for obtaining the abnormality degree includes: For each dimension, the difference between the dimension data at the target sampling moment and each other dimension data in the corresponding neighborhood is calculated as the first difference; the first difference between the dimension data at the target sampling moment and all other dimension data in the corresponding neighborhood is accumulated, and negative correlation normalization mapping is performed to obtain the first abnormal feature; Calculate the difference between the target sampling time and the previous adjacent sampling time as the first data difference; Calculate the difference between the dimensional data of the next adjacent sampling time and the target sampling time as the second data difference; Calculating the difference between the first data difference and the second data difference, and normalizing them to obtain a second abnormal feature at the target sampling time; Perform negative correlation normalization mapping on the dimensional data at the target sampling moment as the third abnormal feature; The product of the first abnormal feature, the second abnormal feature and the third abnormal feature is calculated to obtain the abnormality degree of each dimension data at the target sampling time.

[0006] Furthermore, the method for obtaining the local distribution association includes: For each dimension, within the neighborhood of the dimension data at the target sampling time, calculate the percentile of the dimension data at the target sampling time; calculate the difference in percentiles between the two dimension data at the target sampling time as the first difference; Calculate the mean of the percentile differences between the two dimensional data corresponding to each other sampling time in the neighborhood as the first difference mean; The difference between the first difference and the first difference mean is calculated, and a negative correlation mapping is performed as a local distribution association between the two dimensional data at the target sampling time.

[0007] Furthermore, the method for obtaining the connectivity includes: At the target sampling time, the difference in the degree of abnormality between the two dimensional data is calculated, and negative correlation mapping is performed to obtain the first correlation degree; The product between the first correlation degree and the local distribution correlation is calculated, and negative correlation mapping is performed to obtain the connection between the two dimensional data at the target sampling time.

[0008] Furthermore, the method for obtaining the modified score coefficient includes: Calculate the product of the abnormality of the power efficiency data at the target sampling moment and the correlation between the two dimensional data, and normalize them as the normalized value; calculate the sum of the normalized value and the preset constant to obtain the corrected score coefficient of the power efficiency data at the target sampling moment, and the preset constant is a positive number greater than or equal to 1.

[0009] Furthermore, the method for obtaining the initial score value includes: The Z-Score method is used to obtain the initial score value of the electrical efficiency data at the target sampling time.

[0010] Furthermore, the method for obtaining the modified score value includes: At the target sampling time, the product of the initial score value and the correction score coefficient is calculated to obtain the correction score value of the power efficiency data at the target sampling time.

[0011] Furthermore, the filtering out abnormal electrical efficiency data according to the corrected score value includes: If the corrected score value of the power efficiency data at the sampling moment is less than or equal to the preset abnormal score threshold, the power efficiency data at the sampling moment is abnormal power efficiency data.

[0012] Furthermore, the preset abnormality score threshold is -2.

[0013] The present invention also proposes an energy-saving and environmental protection detection system for a green building, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any one of the steps of the energy-saving and environmental protection detection method for a green building.

[0014] The present invention has the following beneficial effects: In order to more accurately analyze the data state at the sampling moment, the present invention selects any sampling moment as the target sampling moment; according to the change trend of all data of the corresponding dimension within the neighborhood range of each dimensional data at the target sampling moment, the abnormality degree of each dimensional data at the target sampling moment is obtained, which data points deviate from the normal mode can be accurately identified, thereby improving the accuracy of abnormality detection; according to the difference in the abnormality degree between the two dimensional data at the target sampling moment, and the local distribution association between the two dimensional data, the connection between the two dimensional data at the target sampling moment is obtained, and the change of the electric efficiency data over time is obtained, so as to understand the overall trend of the energy-saving and environmental protection performance of green buildings; Further obtaining the correction score coefficient of the electric efficiency data at the target sampling moment can adjust the deviation in the electric efficiency data to make it closer to the true value, thereby improving the accuracy of the analysis; correcting the initial score value of the corresponding electric efficiency according to the correction score coefficient of the electric efficiency data at the target sampling moment, obtaining the correction score value of the electric efficiency data at the target sampling moment, more truly reflecting the electric efficiency performance of the building at the target sampling moment, and improving the accuracy of abnormal data detection; changing the target sampling moment, obtaining the correction score value of the electric efficiency data at each sampling moment, obtaining the change of the electric efficiency data over time, thereby understanding the overall trend of the energy-saving and environmental protection performance of the green building; screening out abnormal electric efficiency data. The present invention obtains an accurate score value for each electric efficiency data, improves the accuracy of abnormal electric efficiency data detection, and optimizes energy-saving and environmental protection detection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 A flow chart of a green building energy-saving and environmental protection detection method provided by one embodiment of the present invention; Figure 2 A waveform diagram of collecting electrical efficiency data and power factor data in a green building provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the energy-saving and environmental protection detection method and system for green buildings proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0019] The specific scheme of the energy-saving and environmental protection detection method and system for green buildings provided by the present invention is described in detail below with reference to the accompanying drawings.

[0020] See also Figure 1 , which shows a flow chart of a green building energy-saving and environmental protection detection method provided by an embodiment of the present invention, and the specific method includes: Step S1: Acquire multi-dimensional data of the green building at each sampling moment in a sampling time period; the multi-dimensional data includes electrical efficiency data and power factor data.

[0021] In the embodiment of the present invention, a green building refers to a building that saves resources to the maximum extent, protects the environment and reduces pollution throughout its entire life cycle, provides people with healthy, applicable and efficient use space, and coexists harmoniously with nature; in order to formulate a reasonable energy management strategy, deeply understand the power consumption of the building and the efficiency of electrical equipment, and discover energy-saving potential; first, in the same time period, use an electric energy monitor and a power factor controller to collect data, and obtain multi-dimensional data of the green building at each sampling moment in the sampling time period; the multi-dimensional data includes electric efficiency data and power factor data; such as Figure 2 , a waveform diagram for collecting electrical efficiency data and power factor data of green buildings is given.

[0022] It should be noted that, in one embodiment of the present invention, the method for acquiring electricity efficiency data is: identifying and recording each electrical equipment in the building, including lamps, air conditioners, office equipment, kitchen equipment, etc., and recording the model, rated power and other information of each electrical equipment; using an electric energy monitor to measure and record the electricity consumption of each electrical equipment over a period of time, and calculating the energy efficiency ratio of each electrical equipment through the electricity consumption data and the rated power of the electrical equipment, which is used to evaluate the electrical efficiency of the green building.

[0023] In one embodiment of the present invention, in order to facilitate subsequent data processing, an analog-to-digital conversion device is used to digitally convert the electrical efficiency data and power factor data in the green building to obtain the corresponding data value at each sampling moment.

[0024] It should be noted that, in one embodiment of the present invention, the collection time period is 10 minutes, and the interval between sampling moments is one second; in other embodiments of the present invention, the collection time period and the interval between the collection moments can be set according to specific circumstances, which is not limited or elaborated here.

[0025] Step S2: Select any sampling moment as the target sampling moment; obtain the abnormality degree of each dimensional data at the target sampling moment according to the change trend of all data of the corresponding dimension within the neighborhood range of each dimensional data at the target sampling moment; obtain the connection between the two dimensional data at the target sampling moment according to the difference in the abnormality degree between the two dimensional data at the target sampling moment, and the local distribution association between the two dimensional data; obtain the correction score coefficient of the power efficiency data at the target sampling moment according to the abnormality degree of the power efficiency data at the target sampling moment and the connection between the two dimensional data.

[0026] Data at different sampling moments may show different characteristics. Choosing a specific sampling moment as the target can capture the local change trend at the sampling moment and analyze the data status at the sampling moment more accurately. By analyzing the data changes in the neighborhood range of the data at the target sampling moment, the dynamic change process of the data can be understood, so as to more accurately evaluate its abnormality. The lower the data on each dimension, the greater the corresponding energy consumption in the building, the less energy-saving and environmentally friendly it is, and the more abnormal it is. Select any sampling moment as the target sampling moment. According to the change trend of all data of the corresponding dimension in the neighborhood range of each dimension data at the target sampling moment, the abnormality degree of each dimension data at the target sampling moment is obtained.

[0027] Preferably, in one embodiment of the present invention, the method for obtaining the abnormality degree includes: For each dimension, the difference between the dimension data at the target sampling moment and each other dimension data in the corresponding neighborhood is calculated as the first difference; the first difference between the dimension data at the target sampling moment and all other dimension data in the corresponding neighborhood is accumulated, and negative correlation normalization mapping is performed to obtain the first abnormal feature; Calculate the difference of the dimensional data between the target sampling moment and the previous adjacent sampling moment as the first data difference; calculate the difference of the dimensional data between the next adjacent sampling moment and the target sampling moment as the second data difference; calculate the difference between the first data difference and the second data difference, and normalize them to obtain the second abnormal feature at the target sampling moment; The dimensional data at the target sampling time is negatively correlated and normalized to be used as the third abnormal feature; the product of the first abnormal feature, the second abnormal feature and the third abnormal feature is calculated to obtain the abnormal degree of each dimensional data at the target sampling time. In one embodiment of the present invention, for any dimension, the abnormal degree is expressed as: ; in, Indicates The degree of abnormality of the dimensional data at each target sampling moment; Indicates Dimensional data at each target sampling moment; Indicates The target sampling time is within the neighborhood of the dimensional data. Other dimension data; Indicates The difference between the dimension data of a target sampling moment and the previous adjacent sampling moment; Indicates the difference in dimensional data between the next adjacent sampling time and the target sampling time; Indicates The number of other dimensional data within the neighborhood of the dimensional data at the target sampling time; represents the normalization function; Represents an exponential function with a natural constant as its base.

[0028] In the formula for the degree of abnormality, Represents the difference between the first data difference and the second data difference. If the second data difference is smaller than the first data difference, the dimensional data at the target sampling time cannot be the minimum value, and the possibility of being abnormal is smaller, that is, the smaller the difference between the first data difference and the second data difference, the smaller the second abnormal feature; if the second data difference is larger than the first data difference, the dimensional data at the target sampling time is more likely to be the minimum value, and the data at the corresponding sampling time is more abnormal, that is, the larger the difference between the first data difference and the second data difference, the larger the second abnormal feature; through an exponential function with a natural constant as the base, Perform negative correlation normalization mapping, Represents the first difference. If the value of the dimensional data at the target sampling moment is not the smallest compared with the dimensional data at other sampling moments, the possibility that the corresponding target sampling moment is abnormal dimensional data is smaller, that is, the larger the first difference, the smaller the first abnormal feature; on the contrary, if the value of the dimensional data at the target sampling moment is smaller than that of the dimensional data at other sampling moments, the corresponding target sampling moment is more likely to be abnormal dimensional data, that is, the smaller the first difference, the larger the first abnormal feature; the smaller the dimensional data at the target sampling moment, the larger the third abnormal feature, and the greater the degree of abnormality. It should be noted that, in one embodiment of the present invention, in each dimension, the neighborhood range of the dimensional data at the target sampling time is to select 100 other dimensional data with the dimensional data at the target sampling time as the center to form a neighborhood range. If the central dimensional data has less than half of the selected data on one side, it is necessary to fill in the insufficient data on the other side to form a complete neighborhood range. In one embodiment of the present invention, the size of the neighborhood range can be set according to the specific situation, which is not limited or elaborated here.

[0029] Since noise data caused by environmental factors will be generated during data collection, and the numerical performance is similar to that of abnormal power efficiency data, it is not accurate enough to distinguish the noise data of power efficiency only by the abnormal degree of a single dimension; in the actual building power consumption process, there is a certain connection between power factor data and power efficiency data, such as Figure 2 , when the power factor data is higher, there is often also higher power efficiency data, so the correlation between the two dimensional data is comprehensively considered; the difference in the degree of abnormality between the two dimensional data is analyzed to determine whether it is caused by a single dimension or the result of multiple dimensions working together. The smaller the difference, the greater the connection between the two dimensional data; through the local distribution association between the two dimensional data, the potential relationship and pattern between the dimensional data at the target sampling moment can be indicated, which helps to deeply understand the internal structure and characteristics of the data. Therefore, according to the difference in the degree of abnormality between the two dimensional data at the target sampling moment, and the local distribution association between the two dimensional data, the connection between the two dimensional data at the target sampling moment is obtained.

[0030] Preferably, in one embodiment of the present invention, the method for obtaining local distribution association includes: Percentiles can help us understand the relative position of a piece of data within its overall range, thereby more comprehensively evaluating the characteristics of the data and providing a standardized measurement method that makes comparisons between data of different dimensions more intuitive and fair.

[0031] For each dimension, within the neighborhood of the dimension data at the target sampling time, calculate the percentile of the dimension data at the target sampling time; calculate the difference in percentiles between the two dimension data at the target sampling time as the first difference; The mean of the percentile differences between the two dimensional data corresponding to each other sampling time in the neighborhood is calculated as the first difference mean; the difference between the first difference and the first difference mean is calculated, and a negative correlation mapping is performed as the local distribution association between the two dimensional data at the target sampling time. In one embodiment of the present invention, the formula for the local distribution association is expressed as: ; in, Indicates The local distribution correlation between the two-dimensional data at the target sampling time; Indicates Power efficiency data at target sampling time percentile of Indicates Power factor data at target sampling time percentile of Indicates Power efficiency data at target sampling time In the neighborhood of Percentiles of other dimension data; Indicates Power factor data at target sampling time In the neighborhood of Percentiles of other dimension data; Indicates The number of other dimensional data within the neighborhood of the dimensional data at the target sampling time; Represents an exponential function with a natural constant as its base.

[0032] In the formula for local distribution correlation, Perform negative correlation mapping, Indicates the first difference between the two dimensional data at the target sampling time. The smaller the first difference, the closer the percentiles between the two dimensional data are to the same distribution position, and the more consistent the change trend is. On the contrary, the greater the difference in the change trend is. It represents the mean of the first differences between the two dimensional data corresponding to other sampling moments in the neighborhood, that is, the average change trend between the two dimensional data in the neighborhood; the smaller the difference between the first difference and the mean of the first differences, the smaller the first difference between the two dimensional data at the target sampling moment is relative to the mean of the first difference, the more likely it is that the dimensional data at the target sampling moment have the same level of distribution, and the greater the local distribution correlation.

[0033] It should be noted that, in one embodiment of the present invention, in each dimension, the method for obtaining the percentile of dimensional data is: for the neighborhood range of each dimensional data, sort them in ascending order, calculate the ratio of the position number of each dimensional data to the number of all dimensional data in the neighborhood range, and obtain the percentile.

[0034] Preferably, in one embodiment of the present invention, the method for obtaining connectivity includes: At the target sampling time, the difference in the degree of abnormality between the two dimensional data is calculated, and negative correlation mapping is performed to obtain a first correlation degree; the product between the first correlation degree and the local distribution correlation is calculated, and negative correlation mapping is performed to obtain the connection between the two dimensional data at the target sampling time. In one embodiment of the present invention, the formula for the connection is expressed as: ; in, Indicates The connection between the two dimensions of data at each target sampling moment; Indicates the local distribution association within the corresponding neighborhood range between dimensional data at the target sampling time; Indicates Power efficiency data at target sampling time The degree of abnormality; Indicates Power factor data at target sampling time The degree of abnormality; Indicates the adjustment parameter.

[0035] In the relational formula, Indicates The smaller the difference in the corresponding abnormality degree between the two dimensional data at the target sampling time, the stronger the connection between the power efficiency data and the power factor data at the target sampling time; the greater the local distribution correlation between the two dimensional data at the target sampling time, the stronger the connection.

[0036] It should be noted that in order to avoid the same abnormality between dimensional data, the formula is 0. In one embodiment of the present invention, the adjustment parameter is 0.01; in other embodiments of the present invention, the size of the adjustment parameter can be specifically set according to the implementation personnel, and is not limited or elaborated here.

[0037] In order to comprehensively reflect the comprehensive performance of the electric efficiency data and reveal the mutual influence between different dimensions; the abnormality degree of the data and the connection between the two dimensional data are comprehensively considered, so that the correction score coefficient can adjust the deviation in the electric efficiency data and more accurately reflect the actual electric efficiency performance of the building at the target sampling time; therefore, according to the abnormality degree of the electric efficiency data at the target sampling time and the connection between the two dimensional data, the correction score coefficient of the electric efficiency data at the target sampling time is obtained.

[0038] Preferably, in one embodiment of the present invention, the method for obtaining the modified score coefficient includes: The product of the abnormality of the power efficiency data at the target sampling time and the connection between the two dimensional data is calculated and normalized as a normalized value; the sum of the normalized value and a preset constant is calculated to obtain a correction score coefficient of the power efficiency data at the target sampling time, and the preset constant is a positive number greater than or equal to 1. In one embodiment of the present invention, the formula for the correction score coefficient is expressed as: ; in, Indicates Power efficiency data at target sampling time Correction score coefficient of Indicates Power efficiency data at target sampling time The degree of abnormality; Indicates The connection between the two dimensions of data at each target sampling moment; Indicates a preset constant; Represents the normalization function.

[0039] In the formula for correcting the score coefficient, The greater the connection between the two-dimensional data at the target sampling moment, the smaller the possibility that the electrical efficiency data at the corresponding moment is a noise data point, and the more likely it is that the abnormality is caused by the actual abnormality. The greater the abnormality of the power efficiency data at a target sampling moment, the more abnormal the power efficiency data at the corresponding sampling moment is, and the more the score coefficient needs to be increased to make it easier to be judged as abnormal data.

[0040] It should be noted that, in one embodiment of the present invention, in order to more clearly highlight the relative size and change trend of the data, the preset constant can be set to 1, so that the value range of the modified score coefficient is within In other embodiments of the present invention, the size of the preset constant can be set according to the specific situation and is not limited or elaborated here.

[0041] Step S3: correct the initial score value of the corresponding power efficiency data according to the correction score coefficient of the power efficiency data at the target sampling time to obtain the corrected score value of the power efficiency data at the target sampling time; change the target sampling time to obtain the corrected score value of the power efficiency data at each sampling time.

[0042] The initial score value is to convert the electric efficiency data into a comparable and analyzable numerical form, which can more intuitively understand the level of electric efficiency performance at the target sampling time; the correction score coefficient reflects the abnormal and associated information in the dimensional data. By introducing the correction score coefficient, the initial score value can be calibrated to more realistically reflect the electric efficiency performance of the building at the target sampling time and improve the accuracy of abnormal data detection. Therefore, according to the correction score coefficient of the electric efficiency data at the target sampling time, the initial score value of the corresponding electric efficiency data is corrected to obtain the corrected score value of the electric efficiency data at the target sampling time.

[0043] Preferably, in one embodiment of the present invention, the method for obtaining the initial score value includes: The Z-Score method converts the raw data into a relative standard score value by standardizing the data, so as to facilitate comparison between different data distributions and improve the data comparability. The Z-Score method is used for the electrical efficiency data to obtain the initial score value of the electrical efficiency data at the target sampling time. In one embodiment of the present invention, the formula of the initial score value is expressed as: ; in, Indicates Power efficiency data at target sampling time The initial score value of Indicates Power efficiency data at target sampling time The data value of Represents the data mean of the power efficiency data at all sampling times; Represents the standard deviation of the power efficiency data at all sampling moments.

[0044] In the formula for the initial score value, The smaller the data value of the power efficiency data at a target sampling moment, the greater the deviation from the data mean, the more likely it is that the power efficiency data at the corresponding sampling moment is abnormal, and the smaller the initial score value is, and the more negative it is.

[0045] It should be noted that the specific Z-Score method is a technical means well known to those skilled in the art and will not be described in detail here.

[0046] Preferably, in one embodiment of the present invention, the method for obtaining the modified score value includes: At the target sampling time, the product of the initial score value and the modified score coefficient is calculated to obtain the modified score value of the power efficiency data at the target sampling time. In one embodiment of the present invention, the formula for the modified score value is expressed as: ; in, Indicates Power efficiency data at target sampling time The corrected score value of Indicates Power efficiency data at target sampling time Correction score coefficient of Indicates Power efficiency data at target sampling time The initial score value of .

[0047] In the formula for correcting the score, the initial score is adjusted with the corrected score coefficient. The lower the electric efficiency data is than the data mean at all sampling times, the smaller the initial score is, and the more likely it is that an abnormality exists. The larger the corrected score coefficient is, the smaller the corrected score is, and the more abnormal the electric efficiency data at the corresponding sampling time is.

[0048] In order to have a more comprehensive understanding of the changes in building electrical efficiency and to more accurately locate the time point and possible causes of energy consumption problems, the target sampling time is changed to obtain the corrected score value of the electrical efficiency data at each sampling time.

[0049] Step S4: Filter out abnormal electrical efficiency data according to the corrected score value.

[0050] The corrected score value can accurately reflect the actual situation of the power efficiency data at each sampling moment, identify the actual abnormal data, and avoid being misled by noise data; therefore, the abnormal power efficiency data is screened out according to the corrected score value.

[0051] Preferably, in one embodiment of the present invention, screening out abnormal electrical efficiency data according to the correction score value includes: If the corrected score value of the power efficiency data at the sampling moment is less than or equal to the preset abnormal score threshold, the power efficiency data at the sampling moment is abnormal power efficiency data.

[0052] It should be noted that, in one embodiment of the present invention, the preset abnormality score threshold is -2; in one embodiment of the present invention, the size of the preset abnormality score threshold can be specifically set according to the specific situation, and is not limited or elaborated here.

[0053] Abnormal electrical efficiency data represents energy consumption problems in green buildings. After obtaining the abnormal electrical efficiency data, a detailed energy efficiency assessment is conducted to timely discover potential problems in energy conservation and environmental protection of green buildings, formulate and implement corresponding energy-saving and environmental protection measures, and improve the energy efficiency level of green buildings.

[0054] In summary, the present invention obtains the abnormality of each dimensional data at the target sampling moment according to the change trend of all data of the corresponding dimension within the neighborhood range of each dimensional data at the target sampling moment; obtains the connection between the two dimensional data at the target sampling moment according to the difference in the abnormality between the two dimensional data at the target sampling moment and the local distribution association between the two dimensional data; further obtains the correction score coefficient of the electric efficiency data at the target sampling moment; corrects the initial score value of the corresponding electric efficiency data to obtain the corrected score value of the electric efficiency data at the target sampling moment; obtains the corrected score value of the electric efficiency data at each sampling moment; and screens out abnormal electric efficiency data. The present invention obtains an accurate score value for each electric efficiency data, improves the accuracy of abnormal electric efficiency data detection, and optimizes energy-saving and environmental protection detection results.

[0055] The present invention also proposes an energy-saving and environmental protection detection system for green buildings, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any step of a green building energy-saving and environmental protection detection method.

[0056] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0057] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A green building energy-saving and environmental protection detection method, characterized in that: The method comprises: Acquire multi-dimensional data of the green building at each sampling moment in the sampling time period; the multi-dimensional data includes electrical efficiency data and power factor data; Select any sampling time as the target sampling time; obtain the abnormality of each dimensional data at the target sampling time according to the change trend of all data of the corresponding dimension within the neighborhood range of each dimensional data at the target sampling time; obtain the connection between the two dimensional data at the target sampling time according to the difference in the abnormality between the two dimensional data at the target sampling time and the local distribution association between the two dimensional data; obtain the correction score coefficient of the electric efficiency data at the target sampling time according to the abnormality of the electric efficiency data at the target sampling time and the connection between the two dimensional data; Correcting the initial score value of the corresponding electric efficiency data according to the correction score coefficient of the electric efficiency data at the target sampling time to obtain the corrected score value of the electric efficiency data at the target sampling time; changing the target sampling time to obtain the corrected score value of the electric efficiency data at each sampling time; Abnormal electrical efficiency data is screened out according to the correction score value.

2. The energy-saving and environmental protection detection method for green buildings according to claim 1 is characterized in that: The method for obtaining the abnormality degree includes: For each dimension, the difference between the dimension data at the target sampling moment and each other dimension data in the corresponding neighborhood is calculated as the first difference; the first difference between the dimension data at the target sampling moment and all other dimension data in the corresponding neighborhood is accumulated, and negative correlation normalization mapping is performed to obtain the first abnormal feature; Calculate the difference between the target sampling time and the previous adjacent sampling time as the first data difference; Calculate the difference between the dimensional data of the next adjacent sampling time and the target sampling time as the second data difference; Calculating the difference between the first data difference and the second data difference, and normalizing them to obtain a second abnormal feature at the target sampling time; Perform negative correlation normalization mapping on the dimensional data at the target sampling moment as the third abnormal feature; The product of the first abnormal feature, the second abnormal feature and the third abnormal feature is calculated to obtain the abnormality degree of each dimension data at the target sampling time.

3. The energy-saving and environmental protection detection method for green buildings according to claim 1 is characterized in that: The method for obtaining the local distribution association includes: For each dimension, within the neighborhood of the dimension data at the target sampling time, calculate the percentile of the dimension data at the target sampling time; calculate the difference in percentiles between the two dimension data at the target sampling time as the first difference; Calculate the mean of the percentile differences between the two dimensional data corresponding to each other sampling time in the neighborhood as the first difference mean; The difference between the first difference and the first difference mean is calculated, and negative correlation mapping is performed as the local distribution association between the two dimensional data at the target sampling time.

4. The energy-saving and environmental protection detection method for green buildings according to claim 1 is characterized in that: The method for obtaining the connectivity includes: At the target sampling time, the difference in the degree of abnormality between the two dimensional data is calculated, and negative correlation mapping is performed to obtain the first correlation degree; The product between the first correlation degree and the local distribution correlation is calculated, and negative correlation mapping is performed to obtain the connection between the two dimensional data at the target sampling time.

5. The energy-saving and environmental protection detection method for green buildings according to claim 1 is characterized in that: The method for obtaining the modified score coefficient includes: Calculate the product of the abnormality of the power efficiency data at the target sampling moment and the correlation between the two dimensional data, and normalize them as the normalized value; calculate the sum of the normalized value and the preset constant to obtain the corrected score coefficient of the power efficiency data at the target sampling moment, and the preset constant is a positive number greater than or equal to 1.

6. The energy-saving and environmental protection detection method for green buildings according to claim 1 is characterized in that: The method for obtaining the initial score value includes: The Z-Score method is used to obtain the initial score value of the electrical efficiency data at the target sampling time.

7. The energy-saving and environmental protection detection method for green buildings according to claim 1 is characterized in that: The method for obtaining the modified score value includes: At the target sampling time, the product of the initial score value and the correction score coefficient is calculated to obtain the correction score value of the power efficiency data at the target sampling time.

8. The energy-saving and environmental protection detection method for green buildings according to claim 1 is characterized in that: The step of screening out abnormal electrical efficiency data according to the correction score value comprises: If the corrected score value of the power efficiency data at the sampling moment is less than or equal to the preset abnormal score threshold, the power efficiency data at the sampling moment is abnormal power efficiency data.

9. The energy-saving and environmental protection detection method for green buildings according to claim 8, characterized in that: The preset abnormal score threshold is -2.

10. A green building energy-saving and environmental protection detection system, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the energy-saving and environmental protection detection method for a green building as described in any one of claims 1 to 9 are implemented.

Images