Office area state detection method and device, storage medium and electronic equipment

Abstract

The application discloses an office area state detection method and device, a storage medium and electronic equipment. The method comprises the following steps: in the case that N groups of reference power consumption data generated by N office areas to be detected in a period of time are acquired, each group of reference data is counted to obtain N power consumption characteristic values corresponding to the N office areas to be detected respectively; the N power consumption characteristic values are iteratively classified until an iteration convergence condition is reached; according to classification information generated when the iteration convergence condition is reached, an outlying indication coefficient corresponding to each power consumption characteristic value is determined; in the case that an abnormal power consumption characteristic value is determined from the N statistical data power consumption characteristic values, it is determined that a target office area corresponding to the abnormal power consumption characteristic value is in a specific state. The application solves the technical problem of low detection efficiency of the office area state and can be applied to machine learning and other scenes in the field of artificial intelligence.

Description

Technical Field

[0001] This invention relates to the field of computers, and more specifically, to a method, apparatus, storage medium, and electronic device for detecting the status of an office area. Background Technology

[0002] In recent years, flexible office spaces have emerged in some developed cities. These spaces are created by dividing an entire office building into several small rooms for rent by the room or by the workstation. Given the large area and population of developed cities, it is obviously impractical to rely on manpower to conduct surveys.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This invention provides a method, apparatus, storage medium, and electronic device for detecting the status of an office area, in order to at least solve the technical problem of low detection efficiency of the status of an office area.

[0005] According to one aspect of the present invention, a method for detecting the status of an office area is provided, comprising: acquiring N sets of reference electricity consumption data generated by N office areas to be detected over a period of time; statistically analyzing each set of reference data to obtain N electricity consumption feature values ​​corresponding to the N office areas to be detected, wherein the electricity consumption feature values ​​are used to represent the electricity consumption behavior characteristics of users in the office areas to be detected, and N is an integer greater than or equal to 3; iteratively classifying the N electricity consumption feature values ​​until an iterative convergence condition is reached; and determining the iterative convergence condition based on the condition. The classification information generated by conditional classification determines the outlier indicator coefficient corresponding to each electricity consumption characteristic value. The outlier indicator coefficient is used to indicate the degree to which each electricity consumption characteristic value deviates from the normal electricity consumption characteristic value among the above N electricity consumption characteristic values. Among the above N electricity consumption characteristic values, the number of the above normal electricity consumption characteristic values ​​is greater than the number of abnormal electricity consumption characteristic values. When an abnormal electricity consumption characteristic value is determined from the above N statistical electricity consumption characteristic values, it is determined that the target office area corresponding to the above abnormal electricity consumption characteristic value is in a specific state. The above abnormal electricity consumption characteristic value is the electricity consumption characteristic value whose outlier indicator coefficient reaches the abnormal threshold.

[0006] According to another aspect of the present invention, an office area status detection device is also provided, comprising: a statistics unit, configured to, upon acquiring N sets of reference electricity consumption data generated by N office areas to be detected within a certain period of time, perform statistics on each of the N sets of reference data to obtain N electricity consumption characteristic values ​​corresponding to the N office areas to be detected, wherein the electricity consumption characteristic values ​​are used to represent the electricity consumption behavior characteristics of users in the office areas to be detected, and N is an integer greater than or equal to 3; a classification unit, configured to iteratively classify the N electricity consumption characteristic values ​​until the iterative convergence condition is reached; and a first determination unit, configured to determine based on... When the above-mentioned iterative convergence condition is met, the classification information generated by the classification determines the outlier indicator coefficient corresponding to each electricity consumption feature value. The outlier indicator coefficient is used to indicate the degree to which each electricity consumption feature value is far away from the normal electricity consumption feature value among the above N electricity consumption feature values. The number of normal electricity consumption feature values ​​among the above N electricity consumption feature values ​​is greater than the number of abnormal electricity consumption feature values. The second determining unit is used to determine that the target office area corresponding to the abnormal electricity consumption feature value is in a specific state when an abnormal electricity consumption feature value is determined among the above N statistical electricity consumption feature values. The abnormal electricity consumption feature value is the electricity consumption feature value whose outlier indicator coefficient reaches the abnormal threshold.

[0007] According to another aspect of the present invention, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer program, wherein the computer program is configured to execute the above-described method for detecting the state of the office area when it is run.

[0008] According to another aspect of the present invention, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the above-described method for detecting the state of an office area through the computer program.

[0009] In this embodiment of the invention, when N sets of reference electricity consumption data generated by N office areas to be tested over a period of time are obtained, each set of reference data is statistically analyzed to obtain N electricity consumption feature values ​​corresponding to the N office areas to be tested. These electricity consumption feature values ​​represent the electricity consumption behavior characteristics of users in the office areas to be tested, and N is an integer greater than or equal to 3. The N electricity consumption feature values ​​are iteratively classified until an iterative convergence condition is met. Based on the classification information generated when the iterative convergence condition is met, an outlier indicator coefficient is determined for each electricity consumption feature value. This outlier indicator coefficient represents the degree to which each electricity consumption feature value deviates from the normal electricity consumption feature values ​​among the N electricity consumption feature values. The number of normal electricity consumption feature values ​​is greater than the number of abnormal electricity consumption feature values. When abnormal electricity consumption feature values ​​are identified from the N statistical electricity consumption feature values, the target office area corresponding to the abnormal electricity consumption feature value is determined to be in a specific state. The abnormal electricity consumption feature value is the electricity consumption feature value whose outlier indicator coefficient reaches the abnormal threshold. Using an iterative classification method, normal and abnormal electricity consumption feature values ​​are determined from multiple electricity consumption feature values ​​used to characterize user electricity consumption behavior in the office area. This achieves the technical objective of detecting the state of the office area without sample training, while ensuring a certain level of detection accuracy. This improves the detection efficiency of the office area state and solves the technical problem of low detection efficiency. Attached Figure Description

[0010] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0011] Figure 1 This is a schematic diagram of the application environment of an optional office area status detection method according to an embodiment of the present invention;

[0012] Figure 2 This is a schematic diagram of a flowchart illustrating an optional method for detecting the status of an office area according to an embodiment of the present invention;

[0013] Figure 3 This is a schematic diagram of an optional method for detecting the status of an office area according to an embodiment of the present invention;

[0014] Figure 4 This is a schematic diagram of another optional method for detecting the status of an office area according to an embodiment of the present invention;

[0015] Figure 5This is a schematic diagram of another optional method for detecting the status of an office area according to an embodiment of the present invention;

[0016] Figure 6 This is a schematic diagram of another optional method for detecting the status of an office area according to an embodiment of the present invention;

[0017] Figure 7 This is a schematic diagram of an optional office area status detection device according to an embodiment of the present invention;

[0018] Figure 8 This is a schematic diagram of another optional office area status detection device according to an embodiment of the present invention;

[0019] Figure 9 This is a schematic diagram of another optional office area status detection device according to an embodiment of the present invention;

[0020] Figure 10 This is a schematic diagram of the structure of an optional electronic device according to an embodiment of the present invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0022] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0023] Artificial intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to possess the functions of perception, reasoning, and decision-making.

[0024] Artificial intelligence (AI) is a comprehensive discipline encompassing a wide range of fields, including both hardware and software technologies. Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies primarily include computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0025] Machine learning (ML) is a multidisciplinary field involving probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory. It specifically studies how computers can simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to endow computers with intelligence; its applications span all areas of artificial intelligence. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and instructional learning.

[0026] With the research and advancement of artificial intelligence (AI) technology, AI is being studied and applied in various fields, such as smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, autonomous driving, drones, robots, smart healthcare, and smart customer service. It is believed that with the development of technology, AI will be applied in more fields and play an increasingly important role.

[0027] The solutions provided in this application involve technologies such as machine learning in artificial intelligence, and are specifically illustrated through the following embodiments:

[0028] According to one aspect of the present invention, a method for detecting the status of an office area is provided. Optionally, as an alternative implementation, the above-described method for detecting the status of an office area may be applied to, but is not limited to, [examples of applications]. Figure 1The environment shown may include, but is not limited to, user equipment 102, network 110, and server 112. The user equipment 102 may include, but is not limited to, a display 108, a processor 106, and a memory 104.

[0029] The specific process can be summarized in the following steps:

[0030] Step S102: User equipment 102 acquires N sets of reference power consumption data generated by N office areas to be tested over a period of time, wherein... Figure 1 The office area shown is for testing purposes only;

[0031] In steps S104-S106, user equipment 102 sends N sets of reference power consumption data to server 112 via network 110;

[0032] In step S108, server 112 performs statistical analysis on each of the N sets of reference data through processing engine 116 to obtain N power consumption feature values ​​corresponding to the N office areas to be detected. The N power consumption feature values ​​are then iteratively classified to determine the outlier indicator coefficient corresponding to each power consumption feature value. The detection result is then determined based on the outlier indicator coefficient. The detection result is used to indicate the target office area corresponding to the power consumption feature value whose outlier indicator coefficient reaches the abnormal threshold.

[0033] In steps S110-S112, server 112 sends the detection results to user device 102 via network 110. Processor 106 in user device 102 displays the relevant information of the target office area corresponding to the detection results on display 108 and stores the relevant information of the target office area in memory 104.

[0034] remove Figure 1 Beyond the illustrated example, the above steps can be performed independently by user equipment 102. Specifically, user equipment 102 can perform steps such as statistical analysis of each set of reference data in the N sets of reference data, iterative classification of the N electricity consumption characteristic values ​​to determine the outlier indicator coefficient corresponding to each electricity consumption characteristic value, thereby reducing the processing load on the server. User equipment 102 includes, but is not limited to, handheld devices (such as mobile phones), laptops, desktop computers, and in-vehicle devices. This invention does not limit the specific implementation of user equipment 102.

[0035] Alternatively, as an alternative implementation method, such as Figure 2 As shown, the methods for detecting the status of office areas include:

[0036] S202, after obtaining N sets of reference electricity consumption data generated by N office areas to be tested within a certain period of time, statistical analysis is performed on each set of reference data to obtain N electricity consumption characteristic values ​​corresponding to the N office areas to be tested. The electricity consumption characteristic values ​​are used to represent the electricity consumption behavior characteristics of users in the office areas to be tested, and N is an integer greater than or equal to 3.

[0037] S204, iteratively classify N electricity consumption characteristic values ​​until the iterative convergence condition is met;

[0038] S206, Based on the classification information generated when the iterative convergence condition is met, determine the outlier indicator coefficient corresponding to each electricity consumption characteristic value. The outlier indicator coefficient is used to indicate the degree to which each electricity consumption characteristic value is far away from the normal electricity consumption characteristic value among the N electricity consumption characteristic values. The number of normal electricity consumption characteristic values ​​among the N electricity consumption characteristic values ​​is greater than the number of abnormal electricity consumption characteristic values.

[0039] S208, if an abnormal electricity consumption characteristic value is identified from N statistical electricity consumption characteristic values, determine that the target office area corresponding to the abnormal electricity consumption characteristic value is in a specific state, wherein the abnormal electricity consumption characteristic value is the electricity consumption characteristic value whose outlier indicator coefficient reaches the abnormal threshold.

[0040] Optionally, in this embodiment, the office area status detection method can be applied, but is not limited to, to detecting abnormal office areas occupied by abnormal users. Specifically, considering the abnormal office areas occupied by abnormal users, due to factors such as users adding rooms and restrooms, and some rooms lacking window ventilation, multiple businesses simultaneously accumulate electrical appliance usage. Therefore, the most obvious characteristic of abnormal office areas occupied by abnormal users is that the electricity consumption data generated in these areas will be significantly abnormal compared to other users of the same area. Furthermore, since abnormal office areas occupied by abnormal users are still a minority in the overall environment, or rather, the number of normal office areas occupied by ordinary users is far greater than the number of abnormal office areas occupied by abnormal users, abnormal electricity consumption data naturally belongs to a minority in the overall data. Therefore, by identifying the minority of abnormal electricity consumption data, the corresponding abnormal office areas can be determined. Based on this, the above-mentioned office area status detection method uses an iterative classification approach to characterize the minority of abnormal electricity consumption data with electricity consumption feature values ​​that have a high outlier indicator coefficient. Thus, even without training samples, it achieves the effect of accurately identifying the abnormal office areas corresponding to the minority of abnormal electricity consumption data.

[0041] Optionally, in this embodiment, the office area to be tested may be, but is not limited to, all or a sample of office areas within a preset range. For example, all or a sample of office areas in the target community may be used as the office area to be tested. Alternatively, it may be, but is not limited to, office areas within a preset range whose area similarity reaches an area threshold. For example, considering that target office areas in a specific state are mostly large office areas that are easy to divide into multiple rooms, office areas in the target community with an area of ​​more than 80㎡ may be used as the office area to be tested. Alternatively, it may be, but is not limited to, office areas whose area similarity is within the area range. For example, office areas in the target community with an area of ​​40-50㎡ may be used as the office area to be tested. Alternatively, it may be, but is not limited to, office areas designated for investigation.

[0042] Optionally, in this embodiment, the reference electricity consumption data can be, but is not limited to, data of different granularities. For example, the reference electricity consumption data can be daily granular electricity consumption data, meaning that the electricity consumption data for each day is used as the reference electricity consumption data; or, for another example, the reference electricity consumption data can be monthly granular electricity consumption data, meaning that the electricity consumption data for each month is used as the reference electricity consumption data. In addition, the reference electricity consumption data can also be, but is not limited to, a set of data of different granularities, and is not limited here.

[0043] Optionally, in this embodiment, the reference electricity consumption data may be, but is not limited to, different types of electricity consumption data. Taking daily granular electricity consumption data as an example, it includes at least one of the following: daily total electricity consumption, daily average total electricity consumption, daily peak total electricity consumption, daily off-peak total electricity consumption, and daily peak-hour total electricity consumption. Taking monthly granular electricity consumption data as another example, it includes at least one of the following: monthly total electricity consumption, monthly average total electricity consumption, and monthly off-peak total electricity consumption.

[0044] Optionally, in this embodiment, the electricity consumption characteristic value can be used, but is not limited to, to represent the electricity consumption behavior of users in abnormal office areas that is more representative of the abnormal electricity consumption behavior compared to users in normal office areas. In other words, the electricity consumption characteristic value can be used, but is not limited to, to represent the difference in electricity consumption behavior patterns between users in abnormal office areas and users in normal office areas. For example, users in abnormal office areas may have increased electricity consumption due to the addition of rooms and restrooms, or some rooms lacking window ventilation, resulting in multiple businesses in a small but fully functional area simultaneously using electrical appliances. Therefore, the most obvious characteristic of users in abnormal office areas is that their total electricity consumption is significantly higher than that of other users in the same area. Furthermore, normal users exhibit a certain stability and periodicity in their electricity consumption behavior. Therefore, within a certain time window, users in abnormal office areas and users in normal office areas show significant differences in electricity consumption statistics and the periodicity of changes in electricity consumption curves. Additionally, users in abnormal office areas and normal users differ significantly in personnel mobility, resulting in significant differences in their electricity consumption patterns.

[0045] Optionally, in this embodiment, the electricity consumption feature value may be, but is not limited to, values ​​in multiple dimensions, and the iterative classification may include, but is not limited to, classifying values ​​by magnitude within the same dimension. For example, if the electricity consumption feature value includes values ​​1, 2, 3, and 4 in the same dimension, then the iterative classification may, but is not limited to, classifying the value by magnitude once or multiple times until the iterative convergence condition is met. Optionally, the classification method may be, but is not limited to, binary classification, multi-class classification, etc., and is not further limited.

[0046] To further illustrate this using iterative classification with binary classification as an example, suppose we have test samples a, b, c, and d that share the same dimension. Performing iterative classification on these samples means sequentially performing binary classification operations. For instance, in the first binary classification operation, a is classified as class A, and b, c, and d are classified as class B. Since class A only contains test sample a, it means class A cannot be classified again. Next, a second binary classification operation is performed on class B, classifying b, c, and d into classes C and D. Class C includes b and c, and class D includes d. Similarly, class D cannot be classified again. Finally, a third binary classification operation is performed on class C, classifying b and c into classes E and F. Class E includes b, and class F includes c. Neither class E nor class F can be classified again, indicating that the iterative convergence condition has been met, and the iterative classification ends.

[0047] Optionally, in this embodiment, the classification information may be used, but is not limited to, to represent the classification status of electricity consumption characteristic values ​​in the iterative classification process, such as classification order, classification type, classification pattern, classification frequency, etc.

[0048] Optionally, in this embodiment, the iterative convergence condition may be, but is not limited to, completing the classification for each electricity consumption characteristic value, and / or determining the outlier indicator coefficient for each classification result, and / or the classification order of the iterative classification reaching an order threshold, and / or the number of classifications in the iterative classification reaching a number threshold, etc.

[0049] Optionally, in this embodiment, since the office area in a specific state is only a small part of the overall office area, and the reference power consumption data generated by the office area in a specific state is significantly different from the reference power consumption data generated by the normal office area, the target office area in a specific state can be determined by using, but is not limited to, the Isolation Forest algorithm (iForest), which does not require labeled samples and is highly efficient in training, to calculate the outlier indicator coefficient.

[0050] Optionally, in this embodiment, the Isolation Forest algorithm is a fast anomaly detection method based on ensemble processing. It features linear time complexity and high accuracy, making it a state-of-the-art algorithm suitable for big data processing. Furthermore, the Isolation Forest algorithm is also applicable to anomaly detection in continuous data. Anomalies are defined as "outliers that are easily isolated," which can be understood as sparsely distributed points far from denser groups. Statistically, in the data space, sparsely distributed regions indicate a low probability of data occurring in those regions; therefore, data falling into these regions can be considered anomalies. Thus, the Isolation Forest algorithm no longer describes normal sample points but isolates outliers. These outliers must meet two characteristics: the outlier data occupies a very small amount, and the feature values ​​of the outlier data differ significantly from those of normal data. For example… Figure 3 In the isolated forest algorithm scenario shown, points in the dense region 306 are considered as a group, i.e., normal points 304, while points outside the dense region 306 are considered as points far from the group, i.e., abnormal points 302.

[0051] To further illustrate, in this embodiment, the Isolation Forest algorithm mainly includes two steps: training iForest and calculating the outlier indicator coefficient, as detailed below:

[0052] (1) Training iForest: Sample from the training set (i.e., N electricity consumption features), construct isolated trees, test each isolated tree in the forest, and record the path length. Specific steps are as follows:

[0053] Step 1: Randomly select Ψ points from the training data as subsamples and put them into the root node of an isolated tree;

[0054] Step 2: Randomly specify a dimension and randomly generate a cut point p within the current node's data range. The cut point is generated between the maximum and minimum values ​​of the specified dimension in the current node's data.

[0055] Step 3: The selection of this cutting point generates a hyperplane that divides the current node's data space into two subspaces: points less than p in the currently selected dimension are placed in the left branch of the current node, and points greater than or equal to p are placed in the right branch of the current node.

[0056] Step 4: Recursively follow steps 2 and 3 on the left and right branches of the node, continuously constructing new leaf nodes until the leaf node contains only one piece of data (it cannot be cut any further) or the tree has grown to the set height.

[0057] (2) Calculate the outlier index coefficient: Calculate the outlier index coefficient (anomaly score) for each sample point according to the formula. Since the cutting process is completely random, the ensemble method is needed to converge the results, i.e., repeatedly perform planar cutting from the beginning and then calculate the average value of each cutting result. After obtaining t isolated trees, the training of a single tree is complete. Next, the generated isolated trees can be used to evaluate the test data, i.e., calculate the outlier index coefficient s. For each sample x, the results of each tree need to be calculated comprehensively, and the outlier index coefficient is calculated using the following formula (1):

[0058]

[0059] Where h(x) is the height of x in each tree, and c(Ψ) is the average path length for a given number of samples Ψ, used to standardize the path length h(x) of sample x.

[0060] Finally, the height of each tree is normalized, and an outlier indicator coefficient between 0 and 1 is obtained. This outlier indicator coefficient is then used to determine whether a point is an outlier. If the outlier indicator coefficient is close to 1, it is definitely an outlier; if the outlier indicator coefficient is much less than 0.5, it is definitely not an outlier; if the outlier indicator coefficient scores for all points are around 0.5, then there are likely no outliers in the sample. For example, considering four test samples a, b, c, and d, where a has an outlier indicator coefficient of 0.9, b and c have outlier indicator coefficients of 0.3, and d has an outlier indicator coefficient of 0.2, then a can be identified as the most likely outlier and the test sample that best represents an isolated point.

[0061] It should be noted that, given N sets of reference electricity consumption data generated by N office areas to be tested over a period of time, each set of reference data is statistically analyzed to obtain N electricity consumption feature values. This aims to meticulously depict the differences in electricity consumption behavior patterns between abnormal and normal office areas. By iteratively classifying the N electricity consumption feature values ​​and based on the classification information generated when the iterative convergence condition is met, the system intelligently and efficiently outputs the outlier indicator coefficient corresponding to each electricity consumption feature value. This helps to address the problem of abnormal office areas and improves the accuracy of sampling and investigation.

[0062] To further illustrate, optionally assume that the above-mentioned method for detecting the state of an office area is applied to an application scenario for detecting anomalies in an office area. The execution carrier of this method is as follows: Figure 4As shown, the system includes a detection system 402, a data access module 4022, a data preprocessing module 4024, a feature extraction module 4026, an anomaly detection module 4028, and an alarm display module 4040. The data access module 4022 is primarily responsible for connecting the power input data collected from the power system to the detection system 402. Since some data values ​​may be abnormal during the acquisition, storage, and transmission of power data, the data preprocessing module 4024 is needed to remove and replace these abnormal data. After the data is accessed and preprocessed by the detection system 402, the feature extraction module 4026... Module 4026 extracts daily and monthly electricity consumption behavior features of users for subsequent model modules; the anomaly detection module 4028 trains anomaly detection model based on user-granular user electricity consumption behavior features, and produces the probability value of a user being an anomalous user (i.e., outlier indicator coefficient); the alarm display module 4040 filters users suspected of being anomalous in the office area based on the output probability value of the user being anomalous in the office area, and outputs the distribution of anomalous users in the office area in various dimensions such as region and time on the alarm display platform, and outputs the digital product of office area anomaly risk index.

[0063] According to the embodiments provided in this application, when N sets of reference electricity consumption data generated by N office areas to be tested over a period of time are obtained, each set of reference data is statistically analyzed to obtain N electricity consumption feature values ​​corresponding to the N office areas to be tested. These feature values ​​represent the electricity consumption behavior characteristics of users in the office areas to be tested, and N is an integer greater than or equal to 3. The N electricity consumption feature values ​​are iteratively classified until the iterative convergence condition is met. Based on the classification information generated when the iterative convergence condition is met, an outlier indicator coefficient is determined for each electricity consumption feature value. This outlier indicator coefficient indicates that each electricity consumption feature value is far removed from the normal electricity consumption feature values ​​among the N electricity consumption feature values. The degree of abnormal electricity consumption is such that the number of normal electricity consumption characteristics is greater than the number of abnormal electricity consumption characteristics among N electricity consumption characteristics. When abnormal electricity consumption characteristics are identified among N statistical electricity consumption characteristics, the target office area corresponding to the abnormal electricity consumption characteristics is determined to be in a specific state. Among them, the abnormal electricity consumption characteristics are the electricity consumption characteristics whose outlier indicator coefficient reaches the abnormal threshold. Using an iterative classification method, normal electricity consumption characteristics and abnormal electricity consumption characteristics are identified among multiple electricity consumption characteristics used to characterize the electricity consumption behavior characteristics of users in the office area. This achieves the technical goal of detecting the state of the office area without sample training while ensuring a certain level of detection accuracy, thereby improving the technical effect of improving the detection efficiency of the office area state.

[0064] As an alternative approach, based on the classification information generated when the iterative convergence condition is met, the outlier indicator coefficient corresponding to each electricity consumption characteristic value is determined, including:

[0065] When the first classification information and the second classification information generated when the iterative convergence condition is met are obtained, the first classification information and the second classification information are integrated and calculated to determine the discrete indicator coefficient corresponding to each electricity consumption feature value. The first classification information is the classification information generated during the iterative classification of the first type of data among N electricity consumption feature values, and the second classification information is the classification information generated during the iterative classification of the second type of data among N electricity consumption feature values.

[0066] Optionally, in this embodiment, the electricity consumption feature value may include, but is not limited to, multi-dimensional feature values. Then, when the electricity consumption feature value includes multiple dimensions, the electricity consumption feature value of each dimension is classified separately to obtain their respective classification information.

[0067] It should be noted that, upon obtaining the first and second classification information generated when the iterative convergence condition is met, these two information are integrated and calculated to determine the discrete indicator coefficient corresponding to each electricity consumption characteristic value. The first classification information refers to the classification information generated during the iterative classification of the first class of data among the N electricity consumption characteristic values, and the second classification information refers to the classification information generated during the iterative classification of the second class of data among the N electricity consumption characteristic values. Here, "first" and "second" are used to represent multiple classes, without any numerical limitation.

[0068] Through the embodiments provided in this application, when the first classification information and the second classification information generated by the classification when the iterative convergence condition is met are obtained, the first classification information and the second classification information are integrated and calculated to determine the discrete indicator coefficient corresponding to each electricity consumption feature value. The first classification information is the classification information generated during the iterative classification of the first type of data among N electricity consumption feature values, and the second classification information is the classification information generated during the iterative classification of the second type of data among N electricity consumption feature values. This achieves the purpose of determining the discrete indicator coefficient using more comprehensive classification information and improves the accuracy of the discrete indicator coefficient.

[0069] As an optional approach, the calculation of the first classification information and the second classification information is integrated to determine the discrete indication coefficient corresponding to each electricity consumption characteristic value, including:

[0070] S1, after obtaining the first-order data in the first classification information and the second-order data in the second classification information, calculate the first average of the first-order data and the second-order data, wherein the first-order data is used to represent the classification order of each first-class data in the iterative classification process, and the second-order data is used to represent the classification order of each second-class data in the iterative classification process;

[0071] S2, after obtaining the first-order data in the first classification information and the second-order data in the second classification information, calculate the second average of the first-order data and the second-order data, wherein the first-order data is used to represent the classification order for classifying each first-class data in the iterative classification process, and the second-order data is used to represent the classification order for classifying each second-class data in the iterative classification process;

[0072] S3, integrate the first average value and the second average value to determine the outlier indicator coefficient corresponding to each electricity consumption characteristic value.

[0073] Optionally, in this embodiment, the iterative classification process may include, but is not limited to, multiple generations of classification with a sequence. For example, the first generation of classification is executed first, and if the iterative convergence condition is not met, the next generation of classification (second generation classification) is executed, and so on.

[0074] Optionally, in this embodiment, the classification order may, but is not limited to, represent the generation in which the electricity consumption feature value is classified during the iterative classification process. For example, during the execution of the first generation classification, the first electricity consumption feature value is classified, and if the first electricity consumption feature value is classified, the classification order of the first electricity consumption feature value is determined to be the first generation. Similarly, during the execution of the first generation classification, the second electricity consumption feature value is classified, and if the second electricity consumption feature value is not classified, the second generation classification continues. Assuming that the second electricity consumption feature value is classified in the second generation classification, the classification order of the second electricity consumption feature value is determined to be the second generation.

[0075] Optionally, in this embodiment, integrating the calculation of the first average and the second average may include, but is not limited to, standardizing the second average using the first average.

[0076] It should be noted that, given the first-order data from the first classification information and the second-order data from the second classification information, a first average of the first-order data and the second-order data is calculated. The first-order data represents the classification order of each data point in the first category during the iterative classification process, and the second-order data represents the classification order of each data point in the second category during the iterative classification process. Similarly, given the first-order data from the first classification information and the second-order data from the second classification information, a second average of the first-order data and the second-order data is calculated. The first-order data represents the classification order used to classify each data point in the first category during the iterative classification process, and the second-order data represents the classification order used to classify each data point in the second category during the iterative classification process. The first average and the second average are then integrated to determine the outlier indicator coefficient corresponding to each electricity consumption characteristic value.

[0077] To further illustrate, a scenario where iterative classification is optionally performed on N electricity consumption characteristic values ​​is, for example... Figure 5 As shown, there are five electricity consumption characteristic values: electricity consumption characteristic value 502, electricity consumption characteristic value 504, electricity consumption characteristic value 506, and electricity consumption characteristic value 508. A first-generation classification is then performed on these five electricity consumption characteristic values ​​to obtain a first-generation classification result 510. The first-generation classification result 510 indicates that electricity consumption characteristic value 502 is classified into one category, and electricity consumption characteristic values ​​504, 506, and 508 are classified into another category. It can be seen that electricity consumption characteristic value 502, which is classified into only one category, cannot be further classified. Electricity consumption feature value 502 can be considered as having completed classification, thus determining that the classification information of electricity consumption feature value 502 includes order data (i.e., classification generation 1). For electricity consumption feature values ​​504, 506, and 508, which can still be classified, the next generation of classification is performed until the iterative convergence condition is met, the classification generation n is determined, and this classification generation is used as the order data of all electricity consumption feature values ​​in the current iterative classification process (i.e., classification n times). That is, the classification information of electricity consumption feature value 502 also includes order data (i.e., classification n generations).

[0078] Through the embodiments provided in this application, when the first-order data in the first classification information and the second-order data in the second classification information are obtained, a first average value of the first-order data and the second-order data is calculated, wherein the first-order data is used to represent the classification order of each first-class data in the iterative classification process, and the second-order data is used to represent the classification order of each second-class data in the iterative classification process; when the first-order data in the first classification information and the second-order data in the second classification information are obtained, a second average value of the first-order data and the second-order data is calculated, wherein the first-order data is used to represent the classification order used to classify each first-class data in the iterative classification process, and the second-order data is used to represent the classification order used to classify each second-class data in the iterative classification process; by integrating and calculating the first average value and the second average value, the outlier indicator coefficient corresponding to each electricity consumption characteristic value is determined, thereby achieving the purpose of efficiently integrating multiple classification information to calculate and obtain the outlier indicator coefficient, and realizing the effect of improving the calculation efficiency of the outlier indicator coefficient.

[0079] As an alternative approach, N electricity consumption characteristics are iteratively classified until the iterative convergence condition is met, including:

[0080] S1, if a classification threshold is obtained, classify each of the N electricity consumption characteristic values ​​according to the classification threshold to obtain the electricity consumption characteristic values ​​that reach the classification threshold and the electricity consumption characteristic values ​​that do not reach the classification threshold, and determine the first number of electricity consumption characteristic values ​​that reach the classification threshold and the second number of electricity consumption characteristic values ​​that do not reach the classification threshold.

[0081] S2, if the first quantity and the second quantity are less than or equal to the target threshold, then the convergence condition is determined.

[0082] Optionally, in this embodiment, the classification threshold may be, but is not limited to, a feature value randomly selected within the target threshold range, wherein the target threshold range is composed of the lower limit of the electricity consumption feature value among N electricity consumption feature values ​​and the lower limit of the electricity consumption feature value.

[0083] It should be noted that, given a classification threshold, each of the N electricity consumption characteristics is classified according to the threshold. For example, electricity consumption characteristics that reach the classification threshold are classified into one category, and those that do not reach the classification threshold are classified into another category. If the number of electricity consumption characteristics in a category is greater than the target threshold, then the classification of electricity consumption characteristics in that category continues; conversely, if the number of electricity consumption characteristics in a category is less than or equal to the target threshold, then it is determined that the electricity consumption characteristics in that category have reached the convergence condition.

[0084] To further illustrate, a scenario where iterative classification is optionally performed on N electricity consumption characteristic values ​​is, for example... Figure 5As shown, there are five electricity consumption feature values: electricity consumption feature value 502, electricity consumption feature value 504, electricity consumption feature value 506, and electricity consumption feature value 508. A first-generation classification is then performed on these five feature values ​​to obtain a first-generation classification result 510. The first-generation classification result 510 indicates that electricity consumption feature value 502 is classified into one category, and electricity consumption feature values ​​504, 506, and 508 are classified into another category. It can be seen that electricity consumption feature value 502, which is classified into only one category, cannot be further classified. Therefore, electricity consumption feature value 502 can be considered as having completed classification, thus determining that the classification information of electricity consumption feature value 502 includes order data (i.e., first-generation classification). For electricity consumption feature values ​​504, 506, and 508, which can still be classified, the next-generation classification is performed until all electricity consumption feature values ​​can no longer be classified, which can be considered as reaching the iterative convergence condition.

[0085] Through the embodiments provided in this application, when a classification threshold is obtained, each of the N electricity consumption feature values ​​is classified according to the classification threshold to obtain the electricity consumption feature values ​​that have reached the classification threshold and the electricity consumption feature values ​​that have not reached the classification threshold. A first number of electricity consumption feature values ​​that have reached the classification threshold and a second number of electricity consumption feature values ​​that have not reached the classification threshold are determined. If the first number and the second number are less than or equal to the target threshold, it is determined that the convergence condition has been reached. This achieves the purpose of quickly classifying electricity consumption feature values ​​according to the classification threshold and improves the classification efficiency of electricity consumption feature values.

[0086] As an optional approach, given a classification threshold, each of the N electricity consumption characteristics is classified according to the classification threshold to obtain the electricity consumption characteristics that meet the classification threshold and those that do not, and a first number of electricity consumption characteristics that meet the classification threshold and a second number of electricity consumption characteristics that do not meet the classification threshold are determined, including:

[0087] S1, if the first classification threshold is obtained, classify each of the N electricity consumption feature values ​​according to the first classification threshold to obtain M electricity consumption feature values ​​that reach the first classification threshold and O electricity consumption feature values ​​that do not reach the first classification threshold, where M and O are integers greater than or equal to 1, and the sum of M and O is equal to N.

[0088] S2, when M is not equal to 1, O is equal to 1, and the second classification threshold is obtained, classify each of the M electricity consumption feature values ​​according to the second classification threshold to obtain P electricity consumption feature values ​​that reach the second classification threshold and Q electricity consumption feature values ​​that do not reach the second classification threshold, where P and Q are integers greater than or equal to 1, and the sum of P and Q is equal to M.

[0089] It should be noted that, when the first classification threshold is obtained, each of the N electricity consumption feature values ​​is classified according to the first classification threshold to obtain M electricity consumption feature values ​​that reach the first classification threshold and O electricity consumption feature values ​​that do not reach the first classification threshold, where M and O are integers greater than or equal to 1, and the sum of M and O equals N. When M is not equal to 1, O is equal to 1, and the second classification threshold is obtained, each of the M electricity consumption feature values ​​is classified according to the second classification threshold to obtain P electricity consumption feature values ​​that reach the second classification threshold and Q electricity consumption feature values ​​that do not reach the second classification threshold, where P and Q are integers greater than or equal to 1, and the sum of P and Q equals M. Similarly, if the number of electricity consumption feature values ​​is not 1, it can be considered that the electricity consumption feature value can still be classified, and the classification continues until the number of all electricity consumption feature values ​​is 1, at which point it can be considered that the electricity consumption feature value can no longer be classified, thus determining that the iterative convergence condition has been reached.

[0090] Through the embodiments provided in this application, when a first classification threshold is obtained, each of the N electricity consumption feature values ​​is classified according to the first classification threshold to obtain M electricity consumption feature values ​​that reach the first classification threshold and O electricity consumption feature values ​​that do not reach the first classification threshold, wherein M and O are integers greater than or equal to 1, and the sum of M and O equals N; when M is not equal to 1, O equals 1, and a second classification threshold is obtained, each of the M electricity consumption feature values ​​is classified according to the second classification threshold to obtain P electricity consumption feature values ​​that reach the second classification threshold and Q electricity consumption feature values ​​that do not reach the second classification threshold, wherein P and Q are integers greater than or equal to 1, and the sum of P and Q equals M, thereby achieving the purpose of reasonably classifying electricity consumption feature values ​​and realizing the effect of improving the classification accuracy of electricity consumption feature values.

[0091] As an optional approach, before performing statistical analysis on each of the N sets of reference data, the following steps are included:

[0092] N sets of reference electricity consumption data are preprocessed to correct reference electricity consumption data in a specific state, wherein the specific state includes at least one of the following: electricity consumption data is missing, electricity consumption data is higher than a preset upper limit, electricity consumption data is lower than a preset lower limit, and electricity consumption data is missing or duplicated.

[0093] Optionally, in this embodiment, during the collection, storage, and transmission of electricity consumption data, some electricity consumption data may be abnormal: for example, missing electricity consumption data, electricity consumption data becoming negative, or electricity consumption data being much larger than normal electricity consumption data (e.g., daily electricity consumption of tens of thousands of kWh); it is also possible that some storage formats have changed due to errors (e.g., one electricity consumption data is stored as two different data points); and therefore, the following processing measures can be taken, but are not limited to, for these abnormal electricity consumption data:

[0094] 1. Replace any missing electricity consumption data and any electricity consumption data that has become negative with 0;

[0095] 2. For electricity consumption data that is extremely high, first calculate the average of the user's total electricity consumption data. If a user's electricity consumption data at a certain time exceeds 100 times the total electricity consumption, then replace the abnormal electricity consumption data with 0.

[0096] 3. If a change in storage format results in two electricity consumption data points being stored at a given time point, then randomly select one of the electricity consumption data points as the current electricity consumption data point.

[0097] It should be noted that N sets of reference electricity consumption data are preprocessed to correct reference electricity consumption data in a specific state, wherein the specific state includes at least one of the following: electricity consumption data is missing, electricity consumption data is higher than the preset upper limit, electricity consumption data is lower than the preset lower limit, and electricity consumption data is missing or duplicated.

[0098] The embodiments provided in this application preprocess N sets of reference electricity consumption data to correct reference electricity consumption data in a specific state, wherein the specific state includes at least one of the following: electricity consumption data is missing, electricity consumption data is higher than a preset upper limit, electricity consumption data is lower than a preset lower limit, and electricity consumption data is missing or duplicated. This achieves the purpose of reducing the inaccuracy of calculation results derived from electricity consumption data due to abnormal electricity consumption data, and realizes the effect of improving the accuracy of calculation results derived from electricity consumption data.

[0099] As an optional approach, statistics are performed on each of the N sets of reference data to obtain N electricity consumption characteristic values ​​corresponding to the N office areas to be tested, including at least one of the following:

[0100] S1, Perform the first type of statistics on each of the N sets of reference data to obtain N first power consumption characteristic values ​​corresponding to the N office areas to be tested, wherein the first power consumption characteristic values ​​are used to represent the power consumption characteristics of users in the office areas to be tested;

[0101] S2, perform second-type statistics on each of the N sets of reference data to obtain N second-type power consumption characteristic values ​​corresponding to the N office areas to be tested. The second-type power consumption characteristic values ​​are used to represent the power consumption frequency characteristics of users in the office areas to be tested.

[0102] S3, perform third-class statistics on each of the N sets of reference data to obtain N third-class electricity consumption characteristic values ​​corresponding to the N office areas to be tested. The third-class electricity consumption characteristic values ​​are used to represent the electricity consumption frequency characteristics of users in the office areas to be tested.

[0103] S4. Perform fourth-category statistics on each of the N sets of reference data to obtain N fourth-category electricity consumption characteristic values ​​corresponding to the N office areas to be tested. The fourth-category electricity consumption characteristic values ​​are used to represent the electricity consumption cycle characteristics of users in the office areas to be tested.

[0104] Optionally, in this embodiment, different calculation methods can be performed on the electricity consumption data based on different construction approaches to obtain different electricity consumption behavior characteristics representing users in the office area to be detected. Furthermore, to improve the accuracy of electricity consumption characteristic values, a combination of multiple types of electricity consumption characteristic values ​​can be used to obtain combined electricity consumption characteristic values.

[0105] Optionally, in this embodiment, considering that abnormalities in office areas may lead to building renovations, an increase in the number of residents, an increase in the use of electrical appliances, and higher electricity consumption, the first type of statistics is performed on each of the N sets of reference data to obtain N electricity consumption characteristics of users in the N office areas to be tested. These characteristics include the ranking characteristics of users' average / valley electricity consumption in the corresponding community, the statistical characteristics of electricity consumption level / valley value, the statistical characteristics of users' electricity consumption level / valley / total lighting over the whole year, and the number of users whose monthly electricity consumption level is less than A (e.g., 50) and valley value is less than B (e.g., 10).

[0106] Furthermore, considering the high mobility of personnel in office areas and the vacancy periods in the premises, resulting in relatively large fluctuations in user electricity consumption, a second type of statistical analysis was performed on each of the N sets of reference data to obtain N electricity frequency characteristics for users in the N office areas to be tested, such as the average / valley / total electricity consumption change characteristics in adjacent years and the fluctuation characteristics of adjacent months (e.g., the difference in electricity consumption between January and February).

[0107] In addition, considering that abnormal users in the office area are more likely to be office workers, and office workers use less electricity at night, the difference between peak and off-peak electricity consumption will be greater. Therefore, a second type of statistics is performed on each of the N sets of reference data to obtain N electricity frequency characteristics of users in the N office areas to be tested, such as statistical values ​​of the difference and ratio of peak and off-peak electricity consumption.

[0108] Furthermore, considering that there may be gaps in office area usage but that electricity consumption has a certain periodicity and regularity, a fourth type of statistical analysis was performed on each of the N sets of reference data to obtain N electricity consumption cycle characteristics for users in the N office areas to be tested. These characteristics include the proportion of 0% electricity consumption in a month, the proportion of high electricity consumption clusters (exceeding the relationship between the mean and variance), the similarity features of electricity consumption values ​​in adjacent months, and the comprehensive statistical features of electricity consumption at different time granularities for users.

[0109] It should be noted that, for each of the N sets of reference data, a first type of statistical analysis is performed to obtain N first power consumption characteristic values ​​corresponding to the N office areas to be tested, where the first power consumption characteristic values ​​represent the power consumption characteristics of users in the office areas to be tested; a second type of statistical analysis is performed for each of the N sets of reference data to obtain N second power consumption characteristic values ​​corresponding to the N office areas to be tested, where the second power consumption characteristic values ​​represent the power consumption frequency characteristics of users in the office areas to be tested; a third type of statistical analysis is performed for each of the N sets of reference data to obtain N third power consumption characteristic values ​​corresponding to the N office areas to be tested, where the third power consumption characteristic values ​​represent the power consumption frequency characteristics of users in the office areas to be tested; and a fourth type of statistical analysis is performed for each of the N sets of reference data to obtain N fourth power consumption characteristic values ​​corresponding to the N office areas to be tested, where the fourth power consumption characteristic values ​​represent the power consumption cycle characteristics of users in the office areas to be tested.

[0110] To further illustrate, an alternative could be to combine daily and monthly electricity consumption data to detect the status of the office area. Figure 6 As shown, the specific steps are as follows:

[0111] S602, acquire daily and monthly granular electricity consumption data;

[0112] S604, perform data cleaning on the acquired daily and monthly granular electricity consumption data to remove abnormal data from the acquired daily and monthly granular electricity consumption data;

[0113] S606 performs data statistics and feature extraction on the cleaned daily and monthly granular power consumption data to obtain the corresponding power consumption feature values.

[0114] S608, input the electricity consumption characteristic value into the detection model;

[0115] S610: Obtain the outlier indicator coefficient output by the detection model, and determine whether the office area to be detected is in a specific state based on the outlier indicator coefficient.

[0116] Through the embodiments provided in this application, a first type of statistical analysis is performed on each of the N sets of reference data to obtain N first power consumption characteristic values ​​corresponding to the N office areas to be tested, wherein the first power consumption characteristic values ​​are used to represent the power consumption characteristics of users in the office areas to be tested; a second type of statistical analysis is performed on each of the N sets of reference data to obtain N second power consumption characteristic values ​​corresponding to the N office areas to be tested, wherein the second power consumption characteristic values ​​are used to represent the power consumption frequency characteristics of users in the office areas to be tested; a third type of statistical analysis is performed on each of the N sets of reference data to obtain N third power consumption characteristic values ​​corresponding to the N office areas to be tested, wherein the third power consumption characteristic values ​​are used to represent the power consumption frequency characteristics of users in the office areas to be tested; and a fourth type of statistical analysis is performed on each of the N sets of reference data to obtain N fourth power consumption characteristic values ​​corresponding to the N office areas to be tested, wherein the fourth power consumption characteristic values ​​are used to represent the power consumption cycle characteristics of users in the office areas to be tested. This achieves the goal of comprehensively obtaining power consumption characteristic values ​​and improves the comprehensiveness of obtaining power consumption characteristic values.

[0117] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0118] According to another aspect of the present invention, an office area status detection device is also provided for implementing the above-described office area status detection method. For example... Figure 7 As shown, the device includes:

[0119] The statistical unit 702 is used to perform statistics on each of the N sets of reference power consumption data generated by N office areas to be tested within a certain period of time, so as to obtain N power consumption characteristic values ​​corresponding to the N office areas to be tested. The power consumption characteristic values ​​are used to represent the power consumption behavior characteristics of users in the office areas to be tested, and N is an integer greater than or equal to 3.

[0120] Classification unit 704 is used to iteratively classify N electricity consumption characteristic values ​​until the iterative convergence condition is met.

[0121] The first determining unit 706 is used to determine the outlier indicator coefficient corresponding to each electricity consumption feature value based on the classification information generated when the iterative convergence condition is reached. The outlier indicator coefficient is used to represent the degree to which each electricity consumption feature value is far away from the normal electricity consumption feature value among the N electricity consumption feature values. The number of normal electricity consumption feature values ​​among the N electricity consumption feature values ​​is greater than the number of abnormal electricity consumption feature values.

[0122] The second determining unit 708 is used to determine that the target office area corresponding to the abnormal electricity consumption characteristic value is in a specific state when an abnormal electricity consumption characteristic value is determined from N statistical electricity consumption characteristic values, wherein the abnormal electricity consumption characteristic value is the electricity consumption characteristic value whose outlier indicator coefficient reaches the abnormal threshold.

[0123] Optionally, in this embodiment, the office area status detection device can be applied, but is not limited to, to detecting abnormal office areas occupied by abnormal users. Specifically, considering the abnormal office areas occupied by abnormal users, due to factors such as users adding rooms and bathrooms, not every household having a kitchen, and some rooms lacking window ventilation, multiple businesses simultaneously accumulate the use of electrical appliances. Therefore, the most obvious characteristic of abnormal office areas occupied by abnormal users is that the electricity consumption data generated in these areas will be significantly abnormal compared to other users of the same area. Furthermore, since abnormal office areas occupied by abnormal users are still a minority in the overall environment, or rather, the number of normal office areas occupied by ordinary users is far greater than the number of abnormal office areas occupied by abnormal users, abnormal electricity consumption data naturally belongs to a minority in the overall data. Thus, by identifying the minority of abnormal electricity consumption data, the corresponding abnormal office areas can be determined. Based on this, the aforementioned office area status detection device uses an iterative classification method to characterize the minority of abnormal electricity consumption data with electricity consumption feature values ​​that have a high outlier indicator coefficient. Therefore, even without training samples, it can accurately determine the abnormal office areas corresponding to the minority of abnormal electricity consumption data.

[0124] Optionally, in this embodiment, the office area to be tested may be, but is not limited to, all or a sample of office areas within a preset range. For example, all or a sample of office areas in the target community may be used as the office area to be tested. Alternatively, it may be, but is not limited to, office areas within a preset range whose area similarity reaches an area threshold. For example, considering that target office areas in a specific state are mostly large office areas that are easy to divide into multiple rooms, office areas in the target community with an area of ​​more than 80㎡ may be used as the office area to be tested. Alternatively, it may be, but is not limited to, office areas whose area similarity is within a certain range. For example, office areas in the target community with an area of ​​40-50㎡ may be used as the office area to be tested. Alternatively, it may be, but is not limited to, designated office areas for sampling. For example, if a report is received from a concerned citizen, the reported office area may be subject to targeted testing.

[0125] Optionally, in this embodiment, the reference electricity consumption data can be, but is not limited to, data of different granularities. For example, the reference electricity consumption data can be daily granular electricity consumption data, meaning that the electricity consumption data for each day is used as the reference electricity consumption data; or, for another example, the reference electricity consumption data can be monthly granular electricity consumption data, meaning that the electricity consumption data for each month is used as the reference electricity consumption data. In addition, the reference electricity consumption data can also be, but is not limited to, a set of data of different granularities, and is not limited here.

[0126] Optionally, in this embodiment, the reference electricity consumption data may be, but is not limited to, different types of electricity consumption data. Taking daily granular electricity consumption data as an example, it includes at least one of the following: daily total electricity consumption, daily average total electricity consumption, daily peak total electricity consumption, daily off-peak total electricity consumption, and daily peak-hour total electricity consumption. Taking monthly granular electricity consumption data as another example, it includes at least one of the following: monthly total electricity consumption, monthly average total electricity consumption, and monthly off-peak total electricity consumption.

[0127] Optionally, in this embodiment, the electricity consumption characteristic value can be used, but is not limited to, to represent the electricity consumption behavior of users in abnormal office areas that is more representative of the abnormal electricity consumption behavior compared to users in normal office areas. In other words, the electricity consumption characteristic value can be used, but is not limited to, to represent the difference in electricity consumption behavior patterns between users in abnormal office areas and users in normal office areas. For example, users in abnormal office areas may have increased electricity consumption due to the addition of rooms and restrooms, or some rooms lacking window ventilation, resulting in multiple businesses in a small but fully functional area simultaneously using electrical appliances. Therefore, the most obvious characteristic of users in abnormal office areas is that their total electricity consumption is significantly higher than that of other users in the same area. Furthermore, normal users exhibit a certain stability and periodicity in their electricity consumption behavior. Therefore, within a certain time window, users in abnormal office areas and users in normal office areas show significant differences in electricity consumption statistics and the periodicity of changes in electricity consumption curves. Additionally, users in abnormal office areas and normal users differ significantly in personnel mobility, resulting in significant differences in their electricity consumption patterns.

[0128] Optionally, in this embodiment, the electricity consumption feature value may be, but is not limited to, values ​​in multiple dimensions, and the iterative classification may include, but is not limited to, classifying values ​​by magnitude within the same dimension. For example, if the electricity consumption feature value includes values ​​1, 2, 3, and 4 in the same dimension, then the iterative classification may, but is not limited to, classifying the value by magnitude once or multiple times until the iterative convergence condition is met. Optionally, the classification method may be, but is not limited to, binary classification, multi-class classification, etc., and is not further limited.

[0129] Optionally, in this embodiment, the classification information may be used, but is not limited to, to represent the classification status of electricity consumption characteristic values ​​in the iterative classification process, such as classification order, classification type, classification pattern, classification frequency, etc.

[0130] Optionally, in this embodiment, the iterative convergence condition may be, but is not limited to, completing the classification for each electricity consumption characteristic value, and / or determining the outlier indicator coefficient for each classification result, and / or the classification order of the iterative classification reaching an order threshold, and / or the number of classifications in the iterative classification reaching a number threshold, etc.

[0131] Optionally, in this embodiment, since the office area in a specific state is only a small part of the overall office area, and the reference power consumption data generated by the office area in a specific state is significantly different from the reference power consumption data generated by the normal office area, the target office area in a specific state can be determined by using, but is not limited to, the Isolation Forest algorithm (iForest), which does not require labeled samples and is highly efficient in training, to calculate the outlier indicator coefficient.

[0132] Optionally, in this embodiment, the Isolation Forest algorithm is a fast anomaly detection device based on ensemble processing. It has linear time complexity and high accuracy, making it a state-of-the-art algorithm that meets the requirements of big data processing. Furthermore, the Isolation Forest algorithm is also suitable for anomaly detection in continuous data. Anomalies are defined as "outliers that are easily isolated," which can be understood as sparsely distributed points that are far from denser groups. Statistically, in the data space, sparsely distributed regions indicate that the probability of data occurring in these regions is very low. Therefore, data falling into these regions can be considered anomalies. Thus, the Isolation Forest algorithm no longer describes normal sample points but rather isolates outliers. These outliers must meet two characteristics: the outlier data occupies a very small amount, and the feature values ​​of the outlier data differ significantly from those of the normal data.

[0133] It should be noted that, given N sets of reference electricity consumption data generated by N office areas to be tested over a period of time, each set of reference data is statistically analyzed to obtain N electricity consumption feature values. This aims to meticulously depict the differences in electricity consumption behavior patterns between abnormal and normal office areas. By iteratively classifying the N electricity consumption feature values ​​and based on the classification information generated when the iterative convergence condition is met, the system intelligently and efficiently outputs the outlier indicator coefficient corresponding to each electricity consumption feature value. This helps to address the problem of abnormal office areas and improves the accuracy of sampling and screening.

[0134] For specific implementation examples, please refer to the example shown in the above-described method for detecting the status of the office area; these examples will not be repeated here.

[0135] According to the embodiments provided in this application, when N sets of reference electricity consumption data generated by N office areas to be tested over a period of time are obtained, each set of reference data is statistically analyzed to obtain N electricity consumption feature values ​​corresponding to the N office areas to be tested. These feature values ​​represent the electricity consumption behavior characteristics of users in the office areas to be tested, and N is an integer greater than or equal to 3. The N electricity consumption feature values ​​are iteratively classified until the iterative convergence condition is met. Based on the classification information generated when the iterative convergence condition is met, an outlier indicator coefficient is determined for each electricity consumption feature value. This outlier indicator coefficient indicates that each electricity consumption feature value is far removed from the normal electricity consumption feature values ​​among the N electricity consumption feature values. The degree of abnormal electricity consumption is such that the number of normal electricity consumption characteristics is greater than the number of abnormal electricity consumption characteristics among N electricity consumption characteristics. When abnormal electricity consumption characteristics are identified among N statistical electricity consumption characteristics, the target office area corresponding to the abnormal electricity consumption characteristics is determined to be in a specific state. Among them, the abnormal electricity consumption characteristics are the electricity consumption characteristics whose outlier indicator coefficient reaches the abnormal threshold. Using an iterative classification method, normal electricity consumption characteristics and abnormal electricity consumption characteristics are identified among multiple electricity consumption characteristics used to characterize the electricity consumption behavior characteristics of users in the office area. This achieves the technical goal of detecting the state of the office area without sample training while ensuring a certain level of detection accuracy, thereby improving the technical effect of improving the detection efficiency of the office area state.

[0136] As an alternative solution, for example Figure 8 As shown, the first determining unit 706 includes:

[0137] The calculation module 802 is used to integrate and calculate the first classification information and the second classification information generated when the iterative convergence condition is met, so as to determine the discrete indicator coefficient corresponding to each electricity consumption feature value. The first classification information is the classification information generated during the iterative classification of the first type of data among N electricity consumption feature values, and the second classification information is the classification information generated during the iterative classification of the second type of data among N electricity consumption feature values.

[0138] For specific implementation examples, please refer to the example shown in the above-described method for detecting the status of the office area; these examples will not be repeated here.

[0139] As an alternative, the computing module 802 includes:

[0140] The first calculation submodule is used to calculate the first average value of the first order data and the second order data when the first order data in the first classification information and the second order data in the second classification information are obtained. The first order data is used to represent the classification order of each first type of data in the iterative classification process, and the second order data is used to represent the classification order of each second type of data in the iterative classification process.

[0141] The second calculation submodule is used to calculate the second average of the first-order data and the second-order data when the first-order data in the first classification information and the second-order data in the second classification information are obtained. The first-order data is used to represent the classification order of each first-class data in the iterative classification process, and the second-order data is used to represent the classification order of each second-class data in the iterative classification process.

[0142] The determination submodule is used to integrate the calculation of the first average value and the second average value to determine the outlier indication coefficient corresponding to each electricity consumption characteristic value.

[0143] For specific implementation examples, please refer to the example shown in the above-described method for detecting the status of the office area; these examples will not be repeated here.

[0144] As an alternative solution, for example Figure 9 As shown, classification unit 704 includes:

[0145] The first determining module 902 is used to classify each of the N electricity consumption characteristic values ​​according to the classification threshold when a classification threshold is obtained, so as to obtain the electricity consumption characteristic values ​​that reach the classification threshold and the electricity consumption characteristic values ​​that do not reach the classification threshold, and determine the first number of electricity consumption characteristic values ​​that reach the classification threshold and the second number of electricity consumption characteristic values ​​that do not reach the classification threshold.

[0146] The second determining module 904 is used to determine the convergence condition when the first quantity and the second quantity are less than or equal to the target threshold.

[0147] For specific implementation examples, please refer to the example shown in the above-described method for detecting the status of the office area; these examples will not be repeated here.

[0148] As an alternative approach, the first determining module includes:

[0149] The first classification submodule is used to classify each of the N electricity consumption characteristic values ​​according to the first classification threshold when the first classification threshold is obtained, so as to obtain M electricity consumption characteristic values ​​that reach the first classification threshold and O electricity consumption characteristic values ​​that do not reach the first classification threshold, wherein M and O are integers greater than or equal to 1, and the sum of M and O is equal to N.

[0150] The second classification submodule is used to classify each of the M electricity consumption feature values ​​according to the second classification threshold when M is not equal to 1, O is equal to 1, and the second classification threshold is obtained, so as to obtain P electricity consumption feature values ​​that reach the second classification threshold and Q electricity consumption feature values ​​that do not reach the second classification threshold, where P and Q are integers greater than or equal to 1, and the sum of P and Q is equal to M.

[0151] For specific implementation examples, please refer to the example shown in the above-described method for detecting the status of the office area; these examples will not be repeated here.

[0152] As an alternative approach, it includes:

[0153] The processing unit is used to preprocess the N sets of reference electricity consumption data before performing statistics on each set of reference data in the N sets of reference data, so as to correct the reference electricity consumption data in a specific state, wherein the specific state includes at least one of the following: missing electricity consumption data, electricity consumption data exceeding a preset upper limit, electricity consumption data below a preset lower limit, and missing or duplicate electricity consumption data.

[0154] For specific implementation examples, please refer to the example shown in the above-described method for detecting the status of the office area; these examples will not be repeated here.

[0155] As an alternative, a statistical unit may include at least one of the following:

[0156] The first statistical module is used to perform first-type statistics on each of the N sets of reference data to obtain N first power consumption characteristic values ​​corresponding to the N office areas to be tested. The first power consumption characteristic values ​​are used to represent the power consumption characteristics of users in the office areas to be tested.

[0157] The second statistical module is used to perform second-type statistics on each of the N sets of reference data to obtain N second electricity consumption characteristic values ​​corresponding to the N office areas to be tested. The second electricity consumption characteristic values ​​are used to represent the electricity consumption frequency characteristics of users in the office areas to be tested.

[0158] The third statistical module is used to perform third-type statistics on each of the N sets of reference data to obtain N third-level electricity consumption characteristic values ​​corresponding to the N office areas to be tested. The third-level electricity consumption characteristic values ​​are used to represent the electricity consumption frequency characteristics of users in the office areas to be tested.

[0159] The fourth statistical module is used to perform fourth-type statistics on each of the N sets of reference data to obtain N fourth-type electricity consumption characteristic values ​​corresponding to the N office areas to be tested. The fourth-type electricity consumption characteristic values ​​are used to represent the electricity consumption cycle characteristics of users in the office areas to be tested.

[0160] For specific implementation examples, please refer to the example shown in the above-described method for detecting the status of the office area; these examples will not be repeated here.

[0161] According to another aspect of the present invention, an electronic device for implementing the above-described method for detecting the state of an office area is also provided, such as... Figure 10 As shown, the electronic device includes a memory 1002 and a processor 1004. The memory 1002 stores a computer program, and the processor 1004 is configured to execute the steps of any of the above method embodiments via the computer program.

[0162] Optionally, in this embodiment, the aforementioned electronic device may be located in at least one of a plurality of network devices in a computer network.

[0163] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program:

[0164] S1. Given N sets of reference electricity consumption data generated by N office areas to be tested over a period of time, perform statistics on each set of reference data to obtain N electricity consumption feature values ​​corresponding to the N office areas to be tested. The electricity consumption feature values ​​are used to represent the electricity consumption behavior characteristics of users in the office areas to be tested, and N is an integer greater than or equal to 3.

[0165] S2, iteratively classify the N electricity consumption characteristic values ​​until the iterative convergence condition is met;

[0166] S3, based on the classification information generated when the iterative convergence condition is met, determine the outlier indicator coefficient corresponding to each electricity consumption feature value. The outlier indicator coefficient is used to indicate the degree to which each electricity consumption feature value is far away from the normal electricity consumption feature value among the N electricity consumption feature values. The number of normal electricity consumption feature values ​​among the N electricity consumption feature values ​​is greater than the number of abnormal electricity consumption feature values.

[0167] S4, if an abnormal electricity consumption characteristic value is identified from N statistical electricity consumption characteristic values, determine that the target office area corresponding to the abnormal electricity consumption characteristic value is in a specific state, wherein the abnormal electricity consumption characteristic value is the electricity consumption characteristic value whose outlier indicator coefficient reaches the abnormal threshold.

[0168] Alternatively, as those skilled in the art will understand, Figure 10 The structure shown is for illustrative purposes only. Electronic devices can also be smartphones (such as Android phones, iOS phones, etc.), tablets, PDAs, mobile internet devices (MIDs), PADs, and other terminal devices. Figure 10 This does not limit the structure of the aforementioned electronic devices. For example, the electronic device may also include components that are more... Figure 10 The more or fewer components shown (such as network interfaces, etc.), or having the same Figure 10 The different configurations shown.

[0169] The memory 1002 can be used to store software programs and modules, such as the program instructions / modules corresponding to the office area status detection method and device in this embodiment of the invention. The processor 1004 executes various functional applications and data processing by running the software programs and modules stored in the memory 1002, thereby realizing the aforementioned office area status detection method. The memory 1002 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 1002 may further include memory remotely located relative to the processor 1004, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. Specifically, the memory 1002 may be used, but is not limited to, to store reference power consumption data, power consumption characteristic values, and outlier indicator coefficients, etc. As an example, such as... Figure 10 As shown, the memory 1002 may include, but is not limited to, the statistics unit 702, classification unit 704, first determination unit 706, and second determination unit 708 in the office area status detection device. Furthermore, it may include, but is not limited to, other module units in the office area status detection device, which will not be elaborated upon in this example.

[0170] Optionally, the transmission device 1006 described above is used to receive or send data via a network. Specific examples of the network described above may include wired networks and wireless networks. In one example, the transmission device 1006 includes a Network Interface Controller (NIC), which can be connected to other network devices and routers via a network cable to communicate with the Internet or a local area network. In another example, the transmission device 1006 is a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0171] In addition, the above-mentioned electronic device also includes: a display 1008 for displaying the above-mentioned reference power consumption data, power consumption characteristic values ​​and outlier indicator coefficients, etc.; and a connection bus 1010 for connecting the various module components in the above-mentioned electronic device.

[0172] In other embodiments, the aforementioned terminal device or server can be a node in a distributed system, wherein the distributed system can be a blockchain system, which is a distributed system formed by connecting multiple nodes through network communication. The nodes can form a peer-to-peer (P2P) network, and any form of computing device, such as a server, terminal, or other electronic device, can become a node in the blockchain system by joining this peer-to-peer network.

[0173] According to one aspect of this application, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the aforementioned office area status detection method, wherein the computer program is configured to execute the steps of any of the above-described method embodiments during runtime.

[0174] Optionally, in this embodiment, the computer-readable storage medium described above may be configured to store a computer program for performing the following steps:

[0175] S1. Given N sets of reference electricity consumption data generated by N office areas to be tested over a period of time, perform statistics on each set of reference data to obtain N electricity consumption feature values ​​corresponding to the N office areas to be tested. The electricity consumption feature values ​​are used to represent the electricity consumption behavior characteristics of users in the office areas to be tested, and N is an integer greater than or equal to 3.

[0176] S2, iteratively classify the N electricity consumption characteristic values ​​until the iterative convergence condition is met;

[0177] S3, based on the classification information generated when the iterative convergence condition is met, determine the outlier indicator coefficient corresponding to each electricity consumption feature value. The outlier indicator coefficient is used to indicate the degree to which each electricity consumption feature value is far away from the normal electricity consumption feature value among the N electricity consumption feature values. The number of normal electricity consumption feature values ​​among the N electricity consumption feature values ​​is greater than the number of abnormal electricity consumption feature values.

[0178] S4, if an abnormal electricity consumption characteristic value is identified from N statistical electricity consumption characteristic values, determine that the target office area corresponding to the abnormal electricity consumption characteristic value is in a specific state, wherein the abnormal electricity consumption characteristic value is the electricity consumption characteristic value whose outlier indicator coefficient reaches the abnormal threshold.

[0179] Optionally, in this embodiment, those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0180] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0181] If the integrated units in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in the aforementioned computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause one or more computer devices (which may be personal computers, servers, or network devices, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0182] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0183] In the several embodiments provided in this application, it should be understood that the disclosed client can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between units or modules, and may be electrical or other forms.

[0184] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0185] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0186] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for detecting the status of an office area, characterized in that, include: Given N sets of reference electricity consumption data generated by N office areas to be tested over a period of time, statistics are performed on each of the N sets of reference data to obtain N electricity consumption characteristic values ​​corresponding to the N office areas to be tested. The electricity consumption characteristic values ​​are used to represent the electricity consumption behavior characteristics of users in the office areas to be tested, and N is an integer greater than or equal to 3. The N electricity consumption characteristic values ​​are iteratively classified until the iterative convergence condition is reached. The iterative convergence condition is that each electricity consumption characteristic value is classified, or the outlier indicator coefficient of each classification result is determined, or the classification order of the iterative classification reaches an order threshold, or the number of classifications of the iterative classification reaches a number threshold. The outlier indicator coefficient is used to indicate the degree to which each electricity consumption characteristic value is far away from the normal electricity consumption characteristic value among the N electricity consumption characteristic values. The number of normal electricity consumption characteristic values ​​among the N electricity consumption characteristic values ​​is greater than the number of abnormal electricity consumption characteristic values. Based on the classification information generated when the iterative convergence condition is met, the outlier indicator coefficient corresponding to each electricity consumption characteristic value is determined. If an abnormal electricity consumption feature is identified among the N electricity consumption feature values, the target office area corresponding to the abnormal electricity consumption feature value is determined to be in a specific state. The abnormal electricity consumption feature value is the electricity consumption feature value whose outlier indicator coefficient reaches the abnormal threshold. The specific state includes at least one of the following: missing electricity consumption data, electricity consumption data exceeding a preset upper limit, electricity consumption data falling below a preset lower limit, or missing or duplicate electricity consumption data.

2. The method according to claim 1, characterized in that, The step of determining the outlier indicator coefficient corresponding to each electricity consumption feature value based on the classification information generated when the iterative convergence condition is met includes: When the first classification information and the second classification information generated by the classification are obtained when the iterative convergence condition is met, the first classification information and the second classification information are integrated and calculated to determine the discrete indicator coefficient corresponding to each of the electricity consumption feature values. The first classification information is the classification information generated during the iterative classification of the first type of data among the N electricity consumption feature values, and the second classification information is the classification information generated during the iterative classification of the second type of data among the N electricity consumption feature values.

3. The method according to claim 2, characterized in that, The process of integrating and calculating the first classification information and the second classification information to determine the discrete indication coefficient corresponding to each electricity consumption characteristic value includes: Given the first-order data in the first classification information and the second-order data in the second classification information, calculate the first average value of the first-order data and the second-order data, wherein the first-order data is used to represent the classification order of each first-class data in the iterative classification process, and the second-order data is used to represent the classification order of each second-class data in the iterative classification process. Given the first-order data in the first classification information and the second-order data in the second classification information, calculate the second average of the first-order data and the second-order data, wherein the first-order data is used to represent the classification order for classifying each of the first class data in the iterative classification process, and the second-order data is used to represent the classification order for classifying each of the second class data in the iterative classification process. The first average value and the second average value are integrated and calculated to determine the outlier indicator coefficient corresponding to each electricity consumption characteristic value.

4. The method according to claim 1, characterized in that, The iterative classification of the N electricity consumption characteristics until the iterative convergence condition is met includes: If a classification threshold is obtained, each of the N electricity consumption characteristic values ​​is classified according to the classification threshold to obtain the electricity consumption characteristic values ​​that reach the classification threshold and the electricity consumption characteristic values ​​that do not reach the classification threshold, and a first number of electricity consumption characteristic values ​​that reach the classification threshold and a second number of electricity consumption characteristic values ​​that do not reach the classification threshold are determined. If the first quantity and the second quantity are less than or equal to the target threshold, the convergence condition is determined to have been met.

5. The method according to claim 4, characterized in that, The step of classifying each of the N electricity consumption characteristic values ​​according to the classification threshold when a classification threshold is obtained, to obtain the electricity consumption characteristic values ​​that reach the classification threshold and the electricity consumption characteristic values ​​that do not reach the classification threshold, and determining a first number of electricity consumption characteristic values ​​that reach the classification threshold and a second number of electricity consumption characteristic values ​​that do not reach the classification threshold, includes: If a first classification threshold is obtained, each of the N electricity consumption feature values ​​is classified according to the first classification threshold to obtain M electricity consumption feature values ​​that reach the first classification threshold and O electricity consumption feature values ​​that do not reach the first classification threshold, wherein M and O are integers greater than or equal to 1, and the sum of M and O is equal to N. When M is not equal to 1 and O is equal to 1, and a second classification threshold is obtained, each of the M electricity consumption feature values ​​is classified according to the second classification threshold to obtain P electricity consumption feature values ​​that reach the second classification threshold and Q electricity consumption feature values ​​that do not reach the second classification threshold, where P and Q are integers greater than or equal to 1, and the sum of P and Q is equal to M.

6. The method according to any one of claims 1 to 5, characterized in that, Before performing statistical analysis on each of the N sets of reference data, the following steps are included: The N sets of reference power consumption data are preprocessed to correct the reference power consumption data that is in a specific state.

7. The method according to any one of claims 1 to 5, characterized in that, The statistical analysis of each of the N sets of reference data is performed to obtain N electricity consumption characteristic values ​​corresponding to the N office areas to be tested, including at least one of the following: For each of the N sets of reference data, perform a first type of statistics to obtain N first power consumption characteristic values ​​corresponding to the N office areas to be tested, wherein the first power consumption characteristic values ​​are used to represent the power consumption characteristics of users in the office areas to be tested; A second type of statistics is performed on each of the N sets of reference data to obtain N second electricity consumption characteristic values ​​corresponding to the N office areas to be tested, wherein the second electricity consumption characteristic values ​​are used to represent the electricity consumption frequency characteristics of users in the office areas to be tested; A third type of statistics is performed on each of the N sets of reference data to obtain N third electricity consumption characteristic values ​​corresponding to the N office areas to be tested, wherein the third electricity consumption characteristic values ​​are used to represent the electricity consumption frequency characteristics of users in the office areas to be tested; A fourth type of statistical analysis is performed on each of the N sets of reference data to obtain N fourth electricity consumption characteristic values ​​corresponding to the N office areas to be tested, wherein the fourth electricity consumption characteristic values ​​are used to represent the electricity consumption cycle characteristics of users in the office areas to be tested.

8. A device for detecting the status of an office area, characterized in that, include: The statistical unit is used to perform statistics on each of the N sets of reference electricity consumption data generated by N office areas to be tested within a certain period of time, so as to obtain N electricity consumption characteristic values ​​corresponding to the N office areas to be tested respectively. The electricity consumption characteristic values ​​are used to represent the electricity consumption behavior characteristics of users in the office areas to be tested, and N is an integer greater than or equal to 3. A classification unit is used to iteratively classify the N electricity consumption characteristic values ​​until the iterative convergence condition is reached. The iterative convergence condition is that each electricity consumption characteristic value is classified, or the outlier indicator coefficient of each classification result is determined, or the classification order of the iterative classification reaches an order threshold, or the number of classifications of the iterative classification reaches a number threshold. The outlier indicator coefficient is used to indicate the degree to which each electricity consumption characteristic value is far away from the normal electricity consumption characteristic value among the N electricity consumption characteristic values, and the number of normal electricity consumption characteristic values ​​among the N electricity consumption characteristic values ​​is greater than the number of abnormal electricity consumption characteristic values. The first determining unit is used to determine the outlier indicator coefficient corresponding to each electricity consumption characteristic value based on the classification information generated when the iterative convergence condition is reached. The second determining unit is used to determine that, when an abnormal electricity consumption characteristic value is determined from the N electricity consumption characteristic values, the target office area corresponding to the abnormal electricity consumption characteristic value is in a specific state, wherein the abnormal electricity consumption characteristic value is the electricity consumption characteristic value whose outlier indicator coefficient reaches an abnormal threshold, and the specific state includes at least one of the following: missing electricity consumption data, electricity consumption data exceeding a preset upper limit, electricity consumption data falling below a preset lower limit, and missing or duplicate electricity consumption data.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein the program, when executed, performs the method described in any one of claims 1 to 7.

10. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to execute the method described in any one of claims 1 to 7 through the computer program.

Images