Power load prediction method, device, equipment, storage medium and program product

Abstract

The invention relates to a power load prediction method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: according to a target people number prediction model corresponding to a target building, determining predicted people entering and exiting from the target building corresponding to a target time period; according to the predicted number of people entering and leaving, determining the predicted number of people entering and / or predicted number of people leaving corresponding to at least one user type and the target time period; and according to the predicted number of entering people and / or the predicted number of leaving people, determining a target remaining number of people of the target building corresponding to the target time period, the target remaining number of people being used for performing power load prediction on the target building. By adopting the method, the accuracy of power load prediction of the building can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of power load forecasting, and in particular to a power load forecasting method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] With the participation of virtual power plants, new energy vehicles, low-carbon parks, load operators and other entities in power demand response and power trading under the new power system, accurate and efficient load forecasting will play an increasingly important role in ensuring the stable operation of the power grid, improving the economic benefits of power sales companies, and improving the quality of power consumption for users. In the scenario of predicting the power load of buildings, accurate prediction of users' energy consumption behavior plays an important role in the prediction results.

[0003] However, conventional technologies have the problem of poor accuracy in predicting the power load of buildings. Summary of the invention

[0004] Based on this, it is necessary to provide a power load forecasting method, device, computer equipment, computer readable storage medium and computer program product that can improve the accuracy of power load forecasting of buildings in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for predicting power load. The method comprises:

[0006] Determine the predicted number of people entering and leaving the target building corresponding to the target time period according to the target number prediction model corresponding to the target building;

[0007] Determine, based on the predicted number of people entering and exiting, a predicted number of people entering and / or a predicted number of people leaving corresponding to the target time period for at least one user type;

[0008] The target number of people to be retained in the target building corresponding to the target time period is determined based on the predicted number of people entering and / or the predicted number of people to be left, and the target number of people to be retained is used to predict the power load of the target building.

[0009] In one embodiment, determining the predicted number of people entering and leaving the target time period for at least one user type according to the predicted number of people entering and leaving the target time period includes:

[0010] Clustering the predicted number of people entering and exiting according to each user type to obtain the target predicted number of people entering and exiting corresponding to each user type;

[0011] Determining the predicted number of people entering and exiting according to each of the target predicted number of people entering and exiting and the entry probability function corresponding to each of the user types;

[0012] The predicted number of people leaving is determined according to the target predicted number of people entering and leaving and the survival probability function corresponding to each user type.

[0013] In one embodiment, determining the predicted number of people entering and exiting according to each of the target predicted number of people entering and exiting and the entry probability function corresponding to each of the user types includes:

[0014] For each of the user types, the product of the predicted number of people entering and exiting corresponding to the user type and the entry probability function corresponding to the user type is determined as the predicted number of people entering.

[0015] In one embodiment, determining the predicted number of people leaving according to the predicted number of people entering and leaving the target and the survival probability function corresponding to each user type includes:

[0016] For each of the user types, the product of the predicted number of people entering and leaving the user type and the survival probability function corresponding to the user type is determined as the predicted number of people leaving the user type.

[0017] In one embodiment, determining the target number of people retained in the target building corresponding to the target time period according to the predicted number of people entering and / or the predicted number of people leaving includes:

[0018] For a user type, the difference between the predicted number of people entering and the predicted number of people leaving is determined as the initial number of people retained corresponding to the user type;

[0019] The target number of retained persons is determined based on the initial number of retained persons.

[0020] In one embodiment, the method further comprises:

[0021] Collecting historical user monitoring data of the target building within a historical time period, the historical user monitoring data including the historical number of people entering, the historical entry time, the historical number of people leaving, and the historical leaving time corresponding to each user type and the historical time period;

[0022] The initial number of people prediction model is iteratively trained according to the historical user monitoring data to obtain the target number of people prediction model.

[0023] In one embodiment, the method further comprises:

[0024] For each user type, obtaining the entry time of each user in the user type into the target building and the exit time of each user from the target building;

[0025] According to the entry time, determining the number of people entering corresponding to a plurality of preset moments and the total number of people entering within a preset time period, each of the preset moments being a moment within the preset time period;

[0026] According to the departure time, determining the number of people leaving corresponding to the plurality of preset moments and the total number of people leaving within the preset time period;

[0027] The entry probability function is obtained according to the ratio of each of the entering numbers to the total entering number, and the survival probability function is obtained according to the ratio of each of the leaving numbers to the total leaving number.

[0028] In a second aspect, the present application also provides a power load prediction device. The device comprises:

[0029] A first determination module is used to determine the predicted number of people entering and exiting the target building corresponding to the target time period according to a target number prediction model corresponding to the target building;

[0030] A second determination module is used to determine the predicted number of people entering and leaving the target time period, based on the predicted number of people entering and leaving the target time period, for at least one user type;

[0031] The third determination module is used to determine the target number of people to be retained in the target building corresponding to the target time period according to the predicted number of people entering and / or the predicted number of people to be left, and the target number of people to be retained is used to predict the power load of the target building.

[0032] In a third aspect, the present application further provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the method described in the first aspect are implemented.

[0033] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method described in the first aspect are implemented.

[0034] In a fifth aspect, the present application further provides a computer program product, wherein the computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the method described in the first aspect are implemented.

[0035] In the above-mentioned power load prediction method, device, computer equipment, storage medium and computer program product, the terminal determines the predicted number of people entering and leaving the target building corresponding to the target time period based on the target number prediction model corresponding to the target building, and then determines the predicted number of people entering and leaving the target building and / or the predicted number of people leaving the target building according to the predicted number of people entering and leaving the target building, and then determines the target number of people remaining in the target building corresponding to the target time period according to the predicted number of people entering and leaving the target building, and the target number of people remaining can be used to predict the power load of the target building. In the power load prediction method, since the target number of people prediction model takes into account the characteristics of the retention behavior of different user types related to the target building, an accurate predicted number of people entering and leaving the target building can be obtained. Therefore, the predicted number of people entering and leaving the target building and / or the predicted number of people leaving the target building can be accurately determined according to the predicted number of people entering and leaving the target building, thereby improving the accuracy of the target number of people remaining in the target building according to the predicted number of people entering and leaving the target building, and then the target number of people remaining can be used to accurately predict the power load of the target building, thereby improving the accuracy of the power load prediction of the building. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0037] FIG1 is a diagram of an application environment of a power load prediction method according to an embodiment;

[0038] FIG2 is a schematic diagram of a flow chart of a method for predicting power load in one embodiment;

[0039] FIG3 is a schematic diagram of a flow chart of step 202 in another embodiment;

[0040] FIG4 is a schematic diagram of a flow chart of step 203 in another embodiment;

[0041] FIG5 is a schematic diagram of a flow chart of a method for predicting power load in another embodiment;

[0042] FIG6 is a schematic flow chart of a method for predicting power load in another embodiment;

[0043] FIG7 is a block diagram of a power load prediction device in one embodiment;

[0044] FIG8 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0046] In the application scenario of predicting the power load of a building, the prediction methods can be divided into three categories: white box method, black box method, and gray box method. The white box method is a forward simulation based on physical meaning. After determining the system structure, state parameters and boundary conditions of the building, a physical model is established to simulate and predict the building energy consumption by solving physical equations. At present, the forward simulation method has developed a variety of detailed simulation tools for building energy consumption, including DOE-2, EnergyPlus, DeST, TRNSYS and other hourly energy consumption simulation calculation tools. The black box method is a data-driven machine learning method. When the input variables and output variables are known, the statistical analysis method is used to solve the various parameters of the model according to the existing data to establish an input-output mapping. Common data-driven models used in the field of building energy consumption prediction include multiple regression models, time series models (such as ARIMA, exponential moving average method), machine learning algorithm models (such as support vector machine regression, decision tree) and artificial neural network models, etc. The gray box method is a combination of forward simulation and machine learning. It uses actual energy consumption data combined with energy consumption physical models to calculate energy consumption. It is a building energy consumption method between the black box method and the white box method. However, due to the lack of standard modeling tools and method guidelines, it is difficult to apply in practice. The above three methods have their own advantages and disadvantages in terms of model accuracy, model complexity, interpretability, data dependence, etc., and the selection can be based on application scenarios and needs.

[0047] However, due to the influence of many factors such as the social attributes, behavioral preferences, weather characteristics, and building conditions of electricity users, the complexity of user electricity consumption characteristics makes load forecasting more difficult. In the field of building energy consumption modeling, the research and application of modeling user behavior and associating it with building energy flow include using Markov chains to model user mobility behavior, defining behavioral patterns such as lighting, window opening and closing, shading, and using electrical appliances; and inferring users' movement and entry and exit behaviors in energy-consuming units by monitoring the positioning of users' mobile devices and socket usage data, thereby establishing an accurate user behavior model. However, the above user modeling methods have the following shortcomings: 1) All users are modeled as a whole, without considering the behavioral differences between different user attributes in the building; 2) Modeling is based on the energy consumption behavior of users inside the building, and the modeling of users' movement behavior in and out of the building is not involved; 3) The amount of calculation is larger; 4) The impact of the timeliness and convenience of data collection on ultra-short-term load forecasting is not considered.

[0048] In view of this, an embodiment of the present application provides a method for predicting power load, which improves the accuracy of predicting the power load of a target building using the target number of retained people by improving the accuracy of predicting the target number of retained people in the target building.

[0049] The power load forecasting method provided in the embodiment of the present application can be applied in the application environment shown in FIG. 1. The terminal 102 communicates with the monitoring device 104 through the network. The data storage system can store the data that the monitoring device 104 needs to process. The data storage system can be integrated on the monitoring device 104, or it can be placed on the cloud or other network servers. The data related to the target building collected by the monitoring device 104 can be sent to the terminal 102, and the terminal 102 can determine the predicted number of people entering and leaving the target building corresponding to the target time period according to the target number prediction model corresponding to the target building, and then, according to the predicted number of people entering and leaving, determine the predicted number of people entering and / or the predicted number of people leaving corresponding to the target time period for at least one user type, and then, according to the predicted number of people entering and / or the predicted number of people leaving, determine the target number of people remaining in the target building corresponding to the target time period, and the target number of people remaining is used to predict the power load of the target building. The terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices and portable wearable devices, and the Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. The portable wearable device can be a smart watch, a smart bracelet, a head-mounted device, etc. The monitoring device 104 may be implemented as a single monitoring device or a monitoring device group consisting of multiple monitoring devices.

[0050] In an exemplary embodiment, as shown in FIG2 , a method for predicting power load is provided, which is described by taking the method applied to the terminal in FIG1 as an example, and includes the following steps:

[0051] Step 201, determining the predicted number of people entering and leaving the target building corresponding to the target time period according to the target number prediction model corresponding to the target building.

[0052] It should be noted that the building in the embodiments of the present application refers to a single building or a group of buildings with similar structures and functions that are under the responsibility of the same operating entity, including but not limited to office buildings, shopping malls, hospitals, hotels, etc. The users who enter and exit the building can be natural persons who have retention behavior in the building, such as office staff, temporary visitors, cleaning staff in office buildings, customers in shopping malls, store operators in shopping malls, hotel catering service staff, etc. Therefore, when predicting the retained users in a building, it is necessary to establish different number of people prediction models for different buildings. The number of people prediction models are used to predict the predicted number of people entering and exiting the building in a future time period based on the number of people entering and exiting the building in a historical time period, and then determine the retained users in the building based on the predicted number of people entering and exiting. Among them, the predicted number of people entering and exiting may include the predicted number of people entering and / or the predicted number of people leaving.

[0053] The target time period refers to a future time period corresponding to the predicted number of people entering and leaving the vehicle. For example, if the current time is 10:00, the target time period may be 11:00-12:00 of the same day, or 08:00-12:00 of the next day. This embodiment does not limit the target time period, and the target time period may be determined according to actual needs. Generally, the target time period may be a short time period in the future.

[0054] In this embodiment, a certain building can be selected as the target building, and the population prediction model corresponding to the target building can be determined as the target population prediction model. Then, the number of people entering and leaving the target building in a historical time period is input into the target population prediction model, and the output result of the target population prediction model is determined as the predicted number of people entering and leaving.

[0055] Step 202: Determine the predicted number of people entering and leaving the target time period according to the predicted number of people entering and leaving the target time period.

[0056] It should be noted that users with different attributes have different retention behaviors in the target building. For example, if the target building is an office building, the retention behavior may be daytime office work, or it may be temporary outing. If the target building is a hotel, the retention behavior may be overnight stay, or it may be dining. Therefore, according to different retention behaviors, users entering and leaving the target building can be divided into different user types, and user types can be used to characterize different retention patterns of users in the target building, so that users entering and leaving the target building can be finely divided, and then the number of people entering and leaving each user type can be predicted separately, improving the accuracy of the prediction results.

[0057] Among them, the predicted number of people entering refers to the number of people of a certain user type entering the target building during the target time period, and the predicted number of people leaving refers to the number of people of a certain user type leaving the target building during the target time period. It is understandable that during the target time period, there may be only a situation where users of a certain user type enter the target building, or there may be only a situation where users of a certain user type leave the target building, or there may be a situation where users of a certain user type both enter the target building and leave the target building. Therefore, the predicted number of people entering and leaving may include the predicted number of people entering and leaving. It is understandable that users entering and leaving the target building can be divided into multiple different user types. For example, the user types corresponding to the target building can be 3, or the user types corresponding to the target building can be 5. This is not limited in this embodiment.

[0058] In this embodiment, the terminal can determine the predicted entering users and predicted leaving users who will enter the target building within the target time period from the users corresponding to the predicted number of entering and exiting users, according to the time when each user enters the target building, or leaves the target building, and then determine the predicted number of entering users corresponding to each user type from the predicted entering users according to the retention behavior corresponding to each user type, and determine the predicted number of leaving users corresponding to each user type from the predicted leaving users according to the retention behavior corresponding to each user type.

[0059] As a possible implementation, the predicted number of people entering and exiting corresponding to each user type is equal to the predicted number of people entering; as another possible implementation, the predicted number of people leaving corresponding to each user type is equal to the predicted number of people leaving.

[0060] Step 203, determining the target number of people to be retained in the target building corresponding to the target time period based on the predicted number of people entering and / or the predicted number of people to be left, the target number of people to be retained being used to perform power load prediction on the target building.

[0061] Among them, the target retention number refers to the number of all users who enter the target building and stay inside the target building during the target time period.

[0062] Optionally, in this embodiment, if the predicted number of people entering and exiting includes the predicted number of people entering, the predicted number of people entering can be determined as the target number of people retained for the target time period; alternatively, if the predicted number of people entering and exiting includes the predicted number of people leaving, the target number of people retained for the target time period can be determined to be 0; alternatively, if the predicted number of people entering and exiting includes the predicted number of people entering and the predicted number of people leaving, the difference between the predicted number of people entering and the predicted number of people leaving can be taken as the new number of people retained, and then the sum of the new number of people retained and the historical number of people retained in the target building before the target time period can be taken as the target number of people retained.

[0063] It is understandable that after obtaining the target number of retained people, the energy consumption of the target building can be predicted based on the target number of retained people, thereby predicting the power load of the target building based on the energy consumption.

[0064] In the above-mentioned power load prediction method, the terminal determines the predicted number of people entering and exiting the target building corresponding to the target time period based on the target number prediction model corresponding to the target building, and then determines the predicted number of people entering and exiting and / or the predicted number of people leaving the target building corresponding to at least one user type and the target time period based on the predicted number of people entering and exiting, and then determines the target number of people remaining in the target building corresponding to the target time period based on the predicted number of people entering and / or the predicted number of people leaving. The target number of people remaining can be used to predict the power load of the target building. In the power load prediction method, since the target number of people prediction model takes into account the characteristics of the retention behavior of different user types related to the target building, an accurate predicted number of people entering and exiting can be obtained. Therefore, the predicted number of people entering and exiting can be accurately determined according to the predicted number of people entering and exiting, thereby improving the accuracy of the target number of people remaining in the target building based on the predicted number of people entering and the predicted number of people leaving, and then the target number of people remaining can be used to accurately predict the power load of the target building, thereby improving the accuracy of the power load prediction of the building.

[0065] In the above scenario of determining the predicted number of people entering and / or leaving corresponding to at least one user type and the target time period based on the predicted number of people entering and leaving, the terminal can determine the predicted number of people entering based on the entry probability function corresponding to each user type, and determine the predicted number of people leaving based on the survival probability function corresponding to each user type. In an exemplary embodiment, as shown in FIG3 , the above step 202 includes:

[0066] Step 301 , clustering the predicted number of people entering and exiting according to each user type to obtain the target predicted number of people entering and exiting corresponding to each user type.

[0067] The target predicted number of in-and-out people refers to the number of in-and-out people predicted for each user type within the target time period. It is understandable that the retention behaviors of users of the same user type in a building are similar. Therefore, after obtaining the predicted number of in-and-out people of the target building according to the target number prediction model, the predicted number of in-and-out people can be clustered, and the number of people corresponding to each clustering result can be determined as the target predicted number of in-and-out people corresponding to each user type.

[0068] Clustering refers to clustering the entire sample set into several classes based on the similarity between the samples, so that the samples belonging to the same class are as similar as possible, and the samples belonging to different classes are obviously different. In this embodiment, the retention behaviors of users belonging to the same clustering result are similar, and the retention behaviors of users belonging to different clustering results are obviously different.

[0069] In this embodiment, target users matching each retention behavior can be screened out from users corresponding to the predicted number of entries and exits based on the retention behavior of each user type, and each target user can be used as a clustering center. Then, all users corresponding to the predicted number of entries and exits can be clustered according to the user type to obtain multiple clustering results, and the number of people corresponding to each clustering result can be used as the target predicted number of entries and exits for each user type.

[0070] Step 302: Determine the predicted number of people entering and exiting according to each target predicted number of people entering and exiting and the entry probability function corresponding to each user type.

[0071] It should be noted that each user type has different retention behaviors and different probabilities of entering the target building in different time periods. Therefore, each user type corresponds to an entry probability function, which can characterize the probability of each user type entering the target building at different times.

[0072] In this embodiment, a user type can be taken as an example to describe the process of determining the initial predicted number of people entering each user type. In this embodiment, a target probability value at a time matching the target time period can be selected from the entry probability function corresponding to the user type, and then the predicted number of people entering corresponding to the user type can be calculated by using the target probability value and the target predicted number of people entering and exiting corresponding to the user type.

[0073] In a possible implementation of step 302, for each user type, the product of the target predicted number of in-and-out people corresponding to the user type and the entry probability function corresponding to the user type may be determined as the predicted number of in-and-out people corresponding to the user type.

[0074] Step 303: Determine the predicted number of people leaving based on the predicted number of people entering and leaving each target and the survival probability function corresponding to each user type.

[0075] It should be noted that each user type has different retention behaviors and different probabilities of leaving the target building in different time periods. Therefore, each user type corresponds to a survival probability function, which can characterize the probability of each user type leaving the target building.

[0076] In this embodiment, a user type can be taken as an example to describe the process of determining the initial predicted number of people leaving for each user type. In this embodiment, a target probability value at a time matching the target time period can be selected from the survival probability function corresponding to the user type, and then the predicted number of people leaving corresponding to the user type can be calculated by using the target probability value and the target predicted number of people entering and leaving the user type.

[0077] In a possible implementation of step 303, for each user type, the product of the target predicted number of in-and-out people corresponding to the user type and the survival probability function corresponding to the user type may be determined as the predicted number of out-and-coming people corresponding to the user type.

[0078] In this embodiment, the terminal can obtain the target predicted number of entries and exits corresponding to each user type by clustering the predicted number of entries and exits according to each user type, and then determine the predicted number of entries based on each target predicted number of entries and exits and the entry probability function corresponding to each user type. Then, determine the predicted number of exits based on each target predicted number of entries and exits and the survival probability function corresponding to each user type. Since the entry probability function and the survival probability function corresponding to each user type can accurately characterize the probability of each user type entering and leaving the target building, the predicted number of entries determined according to the entry probability function and the predicted number of exits determined according to the survival probability function can fully consider the characteristics of the retention behavior of different user types, thereby improving the accuracy of the determined predicted number of entries and the predicted number of exits.

[0079] In the above scenario of determining the target number of people retained in the target building corresponding to the target time period based on the predicted number of people entering and / or the predicted number of people leaving, if there is no number of people retained in the target building before the target time period, the number of people whose predicted number of people leaving is greater than the predicted number of people entering can be determined as the number of people retained corresponding to each user type, thereby determining the target number of people retained based on the number of people retained corresponding to each user type. In an exemplary embodiment, as shown in FIG4 , the above step 203 includes:

[0080] Step 401, the difference between the predicted number of people entering and the predicted number of people leaving is determined as the initial number of people remaining.

[0081] It is understandable that, when the predicted number of people entering and exiting includes both the predicted number of people entering and the predicted number of people leaving, the number of people predicted to leave is less than the predicted number of people entering, which is the number of people retained corresponding to each user type. In this embodiment, the number of people retained corresponding to each user type is determined as the initial number of people retained.

[0082] In this embodiment, the terminal may calculate the difference between the predicted number of people entering and the predicted number of people leaving, and then determine the difference as the initial number of people remaining.

[0083] Step 402, determining the target number of retained persons according to each initial number of retained persons.

[0084] In this embodiment, the sum of the initial retained numbers corresponding to each user type may be determined as the target retained number corresponding to the target building.

[0085] In this embodiment, for a user type, the terminal determines the difference between the predicted number of people entering and the predicted number of people leaving as the initial retained number, and then can determine the target retained number based on each initial retained number. Since the process of calculating the difference between the predicted number of people entering and the predicted number of people leaving by the terminal is relatively simple, the initial retained number can be obtained quickly, and thus the target retained number can be quickly obtained based on each initial retained number.

[0086] The specific process of training the initial number prediction model to obtain the target number prediction model is described in detail below. In an exemplary embodiment, as shown in FIG5 , the method further includes:

[0087] Step 501 , collect historical user monitoring data of the target building in a historical time period, the historical user monitoring data including the historical number of people entering, historical entry time, historical number of people leaving, and historical leaving time corresponding to each user type and historical time period.

[0088] Among them, the historical time period refers to the time between the current time period, and the historical user monitoring data refers to the monitoring data that can represent the users entering and leaving the target building within the historical time period. For example, the parking lot data, background service data, WiFi connection data, equipment switch data, energy consumption data and other time series data of users entering and leaving the target building can be obtained through photography, infrared sensing, and motion sensing online monitoring equipment. Among them, the historical number of entries refers to the number of users entering the target building within the historical time period, the historical entry time refers to the time when the user enters the target building within the historical time period, the historical number of exits refers to the number of users leaving the target building within the historical time period, and the historical exit time refers to the time when the user leaves the target building within the historical time period.

[0089] In this embodiment, the terminal can obtain historical user monitoring data and analyze the historical user monitoring data according to the retention behavior characteristics of each user type to obtain the historical number of entries for each user type in the historical time period, the historical entry time of each user, the historical number of exits, and the historical exit time of each user.

[0090] Step 502, iteratively train the initial number of people prediction model based on historical user monitoring data to obtain a target number of people prediction model.

[0091] The initial number prediction model refers to a neural network model used to predict the number of people. Optionally, the initial number prediction model can be a convolutional neural network model or a recurrent neural network model, which is not limited in this embodiment. It can be understood that after iterative training of the initial number prediction model, the model parameters of the initial number prediction model can be adjusted according to the output results of each training. When the optimal output result is obtained, the initial number prediction model corresponding to the optimal output result can be determined as the target number prediction model.

[0092] In this embodiment, the terminal can use historical user monitoring data as training samples, and use the historical number of people entering, historical entry time, historical number of people leaving, and historical departure time corresponding to each user type and historical time period as standard results, iteratively train the initial number of people prediction model, and compare the output results of each training with the standard results, thereby using the comparison results to calculate the loss function value. When the loss function value tends to a fixed value or reaches a minimum, the initial number of people prediction model of the current iteration number is determined as the target number of people prediction model.

[0093] In this embodiment, first, the terminal collects historical user monitoring data of the target building within the historical time period, and then, based on the historical user monitoring data, the initial number of people prediction model is iteratively trained to obtain a target number of people prediction model. Since the historical user monitoring data includes the historical number of people entering, historical entry time, historical number of people leaving, and historical leaving time corresponding to each user type and the historical time period, the retention behavior characteristics of each user type can be taken into account, and the initial number of people prediction model is trained using the historical user monitoring data. This allows the output result of the target number of people prediction model obtained after training to take into account the retention behavior characteristics of each user type, thereby improving the accuracy of the output result of the target number of people prediction model.

[0094] The process of determining the above-mentioned entry probability function and survival probability function is described in detail below. In an exemplary embodiment, as shown in FIG6 , the above-mentioned method further includes:

[0095] Step 601: for each user type, obtain the entry time of each user in the user type into the target building and the exit time of each user in the user type from the target building.

[0096] It is understandable that the entry probability function and survival probability function corresponding to each user type can be determined according to the relevant data of entering and exiting the target building corresponding to different user types. In this embodiment, a user type can be taken as an example to describe the process of determining the entry probability function and survival probability function corresponding to the user type.

[0097] In this embodiment, for each user type, the terminal can divide the users entering and exiting the target building according to the characteristics of the retention behaviors of multiple user types based on the data of users entering and exiting the target building obtained during the historical time period, thereby obtaining the entry and exit data corresponding to the multiple user types, and then obtain the entry time of each user entering the target building from the entry and exit data corresponding to each user type, and obtain the departure time of each user leaving the target building from the entry and exit data corresponding to each user type.

[0098] Step 602, for each user type, according to the entry time, determine the number of people entering corresponding to multiple preset moments and the total number of people entering within the preset time period, each preset moment being a moment within the preset time period.

[0099] The preset time period refers to the time period for counting the number of users entering and leaving the target building, and the preset time refers to the time when the number of users entering and leaving the target building is counted. The preset time may be multiple times within the preset time period. For example, if the preset time period is 8:00-16:00, the preset time may be 8:00, 9:00, 12:00, 15:00, etc. within the same day.

[0100] In this embodiment, the preset time period can be used as a filtering condition, and the number of users corresponding to all users whose entry time meets the preset time period can be determined as the total number of entries; and the preset time can be used as a filtering condition, and the number of users corresponding to the users whose entry time meets the corresponding preset time can be determined as the number of entries corresponding to each preset time.

[0101] Step 603: for each user type, according to the departure time, determine the number of people leaving corresponding to a plurality of preset moments and the total number of people leaving within a preset time period.

[0102] In this embodiment, the preset time period can be used as a filtering condition, and the number of users corresponding to all users whose leaving time meets the preset time period can be determined as the total number of people who have left; and the preset time can be used as a filtering condition, and the number of users corresponding to the users whose leaving time meets the corresponding preset time can be determined as the number of people who have left corresponding to each preset time.

[0103] Step 604: Obtain an entry probability function according to the ratio of each number of people entering to the total number of people entering, and obtain a survival probability function according to the ratio of each number of people leaving to the total number of people leaving.

[0104] The ratio of each number of people entering to the total number of people entering refers to the ratio of the number of people entering at each preset moment to the total number of people entering at the preset time period, and the entry probability function may include a correspondence between multiple sets of ratios and the corresponding preset moments. The ratio of each number of people leaving to the total number of people leaving refers to the ratio of the number of people leaving at each preset moment to the total number of people leaving at the preset time period, and the survival probability function may include a correspondence between multiple sets of ratios and the corresponding preset moments.

[0105] In this embodiment, the entry probability function can be determined according to the corresponding relationship between the ratio of multiple number of people entering and the total number of people entering and the corresponding preset time, and the survival probability function can be determined according to the corresponding relationship between the ratio of multiple number of people leaving and the total number of people leaving and the corresponding preset time.

[0106] Exemplarily, with each preset moment as the horizontal axis and the ratio corresponding to each preset moment as the vertical axis, a curve is drawn between the ratio of each number of people entering and the total number of people entering and each preset moment, so that the function corresponding to the drawn curve is determined as the entry probability function; and a curve is drawn between the ratio of each number of people leaving and the total number of people leaving and each preset moment, so that the function corresponding to the drawn curve is determined as the survival probability function.

[0107] In this embodiment, for each user type, the terminal obtains the entry time of each user in the user type into the target building and the departure time of the target building, and can determine the number of entries corresponding to multiple preset moments and the total number of entries within the preset time period according to the entry time, each preset moment is a moment within the preset time period, and can determine the number of exits corresponding to multiple preset moments and the total number of exits within the preset time period according to the exit time, so that the entry probability function can be obtained according to the ratio of each number of entries to the total number of entries, and the survival probability function can be obtained according to the ratio of each number of exits to the total number of exits. In this process, since the process of determining the ratio of each number of entries to the total number of entries and the process of determining the ratio of each number of exits to the total number of exits are relatively simple, the efficiency of determining the entry probability function or the survival probability function according to the ratio can be improved.

[0108] To facilitate understanding by those skilled in the art, the power load forecasting method provided by the present application is described in detail below. The method may include:

[0109] S1, collecting historical user monitoring data of the target building in a historical time period, the historical user monitoring data includes the historical number of entry, historical entry time, historical number of exit and historical exit time corresponding to each user type and historical time period.

[0110] S2, based on historical user monitoring data, iteratively train the initial crowd prediction model to obtain a target crowd prediction model corresponding to the target building.

[0111] S3, determining the predicted number of people entering and leaving the target building corresponding to the target time period according to the target number prediction model corresponding to the target building.

[0112] S4, clustering the predicted number of people entering and exiting according to each user type to obtain the target predicted number of people entering and exiting corresponding to each user type.

[0113] S5. For each user type, obtain the entry time of each user in the user type entering the target building and the exit time of each user leaving the target building.

[0114] S6, determining the number of people entering corresponding to a plurality of preset moments and the total number of people entering within a preset time period according to the entry time, wherein each preset moment is a moment within the preset time period.

[0115] S7, determining the number of people leaving corresponding to multiple preset moments and the total number of people leaving within the preset time period according to the departure time.

[0116] S8, obtaining an entry probability function according to the ratio of each number of people entering and the total number of people entering, and obtaining a survival probability function according to the ratio of each number of people leaving and the total number of people leaving.

[0117] S9, for each user type, multiplying the product of the predicted number of in-and-out persons corresponding to the user type and the entry probability function corresponding to the user type to determine the predicted number of in-and-out persons corresponding to the user type.

[0118] S10, for each user type, multiplying the predicted number of people entering and leaving the user type by the survival probability function corresponding to the user type to determine the predicted number of people leaving the user type.

[0119] S11, for each user type, the difference between the predicted number of people entering and the predicted number of people leaving is determined as the initial number of people retained.

[0120] S12, determining the target number of retained persons by summing up the initial number of retained persons.

[0121] It should be noted that for the descriptions in the above S1-S12, reference may be made to the relevant descriptions in the above embodiments, and the effects are similar, so this embodiment will not be repeated here.

[0122] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0123] Based on the same inventive concept, the embodiment of the present application also provides a power load prediction device for implementing the power load prediction method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more power load prediction device embodiments provided below can refer to the limitations of the power load prediction method above, and will not be repeated here.

[0124] In one embodiment, as shown in FIG. 7 , a power load prediction device is provided, including: a first determination module 701, a second determination module 702 and a third determination module 703, wherein:

[0125] The first determination module 701 is used to determine the predicted number of people entering and leaving the target building corresponding to the target time period according to the target number prediction model corresponding to the target building.

[0126] The second determination module 702 is used to determine the predicted number of people entering and leaving corresponding to at least one user type and the target time period according to the predicted number of people entering and leaving.

[0127] The third determination module 703 is used to determine the target number of people to be retained in the target building corresponding to the target time period according to the predicted number of people entering and / or the predicted number of people leaving, and the target number of people to be retained is used to predict the power load of the target building.

[0128] The power load prediction device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0129] In one embodiment, the second determining module 702 includes: an acquiring unit, a first determining unit, and a second determining unit, wherein:

[0130] The acquisition unit is used to perform clustering processing on the predicted number of people entering and exiting according to each user type, and obtain the target predicted number of people entering and exiting corresponding to each user type.

[0131] The first determining unit is used to determine the predicted number of people entering according to the predicted number of people entering and exiting each target and the entry probability function corresponding to each user type.

[0132] The second determining unit is used to determine the predicted number of people leaving according to the predicted number of people entering and leaving each target and the survival probability function corresponding to each user type.

[0133] The power load prediction device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0134] In one embodiment, the first determining unit is specifically configured to:

[0135] For each user type, the product of the predicted number of in-and-out persons corresponding to the user type and the entry probability function corresponding to the user type is determined as the predicted number of in-and-out persons.

[0136] The power load prediction device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0137] In one embodiment, the second determining unit is specifically configured to:

[0138] For each user type, the product of the predicted number of people entering and leaving the user type and the survival probability function corresponding to the user type is determined as the predicted number of people leaving the user type.

[0139] The power load prediction device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0140] In one embodiment, the third determination module 703 includes: a third determination unit and a fourth determination unit, wherein:

[0141] The third determining unit is used to determine the difference between the predicted number of people entering and the predicted number of people leaving as the initial number of people retained.

[0142] The fourth determining unit is used to determine the target number of retained persons according to each initial number of retained persons.

[0143] The power load prediction device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0144] In one embodiment, the above device further includes: a collection module and an acquisition module, wherein:

[0145] The collection module is used to collect historical user monitoring data of the target building within a historical time period. The historical user monitoring data includes the historical number of entries, historical entry time, historical number of exits, and historical exit time corresponding to each user type and historical time period.

[0146] The acquisition module is used to iteratively train the initial number prediction model based on historical user monitoring data to obtain the target number prediction model.

[0147] The power load prediction device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0148] In one embodiment, the above device further includes: a second acquisition module, a fourth determination module, a fifth determination module and a third acquisition module, wherein:

[0149] The second acquisition module is used to acquire, for each user type, an entry time of each user in the user type entering the target building and a departure time of each user in the user type leaving the target building.

[0150] A fourth determination module is used to determine the number of people entering corresponding to a plurality of preset moments and the total number of people entering within a preset time period according to the entry time, each preset moment being a moment within the preset time period;

[0151] A fifth determination module, used to determine the number of people leaving corresponding to a plurality of preset moments and the total number of people leaving within a preset time period according to the departure time;

[0152] The third acquisition module is used to acquire an entry probability function according to the ratio of each number of people entering and the total number of people entering, and to acquire a survival probability function according to the ratio of each number of people leaving and the total number of people leaving.

[0153] The power load prediction device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0154] Each module in the above-mentioned power load prediction device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0155] In an exemplary embodiment, as shown in FIG8 . The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface, the display unit and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for predicting power load is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0156] Those skilled in the art will understand that the structure shown in FIG. 8 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0157] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0158] According to the target number prediction model corresponding to the target building, the predicted number of people entering and leaving the target building corresponding to the target time period is determined;

[0159] Determine, based on the predicted number of people entering and exiting, the predicted number of people entering and / or the predicted number of people leaving corresponding to at least one user type and the target time period;

[0160] Based on the predicted number of people entering and / or predicted number of people leaving, the target number of people retained in the target building corresponding to the target time period is determined, and the target number of people retained is used to predict the power load of the target building.

[0161] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0162] Cluster the predicted number of people entering and exiting according to each user type to obtain the target predicted number of people entering and exiting corresponding to each user type;

[0163] Determine the predicted number of people entering and exiting according to the predicted number of people entering and exiting each target and the entry probability function corresponding to each user type;

[0164] According to the predicted number of people entering and leaving each target and the survival probability function corresponding to each user type, the predicted number of people leaving is determined.

[0165] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0166] For each user type, the product of the predicted number of in-and-out persons corresponding to the user type and the entry probability function corresponding to the user type is determined as the predicted number of in-and-out persons.

[0167] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0168] For each user type, the product of the predicted number of people entering and leaving the user type and the survival probability function corresponding to the user type is determined as the predicted number of people leaving the user type.

[0169] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0170] For a user type, the difference between the predicted number of people entering and the predicted number of people leaving is determined as the initial number of people retained corresponding to the user type;

[0171] Determine the target retention number based on the initial retention number.

[0172] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0173] Collect historical user monitoring data of the target building within a historical time period, the historical user monitoring data including the historical number of people entering, historical entry time, historical number of people leaving, and historical leaving time corresponding to each user type and historical time period;

[0174] According to the historical user monitoring data, the initial number prediction model is iteratively trained to obtain the target number prediction model.

[0175] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0176] For each user type, obtain the entry time of each user in the user type into the target building and the exit time of each user from the target building;

[0177] According to the entry time, determine the number of people entering corresponding to multiple preset moments and the total number of people entering within the preset time period, each preset moment being a moment within the preset time period;

[0178] According to the departure time, determine the number of people leaving corresponding to multiple preset moments and the total number of people leaving within the preset time period;

[0179] The entry probability function is obtained according to the ratio of each number of people entering and the total number of people entering, and the survival probability function is obtained according to the ratio of each number of people leaving and the total number of people leaving.

[0180] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0181] According to the target number prediction model corresponding to the target building, the predicted number of people entering and leaving the target building corresponding to the target time period is determined;

[0182] Determine, based on the predicted number of people entering and exiting, the predicted number of people entering and / or the predicted number of people leaving corresponding to at least one user type and the target time period;

[0183] Based on the predicted number of people entering and / or predicted number of people leaving, the target number of people retained in the target building corresponding to the target time period is determined, and the target number of people retained is used to predict the power load of the target building.

[0184] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0185] Cluster the predicted number of people entering and exiting according to each user type to obtain the target predicted number of people entering and exiting corresponding to each user type;

[0186] Determine the predicted number of people entering and exiting according to the predicted number of people entering and exiting each target and the entry probability function corresponding to each user type;

[0187] According to the predicted number of people entering and leaving each target and the survival probability function corresponding to each user type, the predicted number of people leaving is determined.

[0188] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0189] For each user type, the product of the predicted number of in-and-out persons corresponding to the user type and the entry probability function corresponding to the user type is determined as the predicted number of in-and-out persons.

[0190] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0191] For each user type, the product of the predicted number of people entering and leaving the user type and the survival probability function corresponding to the user type is determined as the predicted number of people leaving the user type.

[0192] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0193] For a user type, the difference between the predicted number of people entering and the predicted number of people leaving is determined as the initial number of people retained corresponding to the user type;

[0194] Determine the target retention number based on the initial retention number.

[0195] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0196] Collect historical user monitoring data of the target building within a historical time period, the historical user monitoring data including the historical number of people entering, historical entry time, historical number of people leaving, and historical leaving time corresponding to each user type and historical time period;

[0197] According to the historical user monitoring data, the initial number prediction model is iteratively trained to obtain the target number prediction model.

[0198] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0199] For each user type, obtain the entry time of each user in the user type into the target building and the exit time of each user from the target building;

[0200] According to the entry time, determine the number of people entering corresponding to multiple preset moments and the total number of people entering within the preset time period, each preset moment being a moment within the preset time period;

[0201] According to the departure time, determine the number of people leaving corresponding to multiple preset moments and the total number of people leaving within the preset time period;

[0202] The entry probability function is obtained according to the ratio of each number of people entering and the total number of people entering, and the survival probability function is obtained according to the ratio of each number of people leaving and the total number of people leaving.

[0203] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0204] According to the target number prediction model corresponding to the target building, the predicted number of people entering and leaving the target building corresponding to the target time period is determined;

[0205] Determine, based on the predicted number of people entering and exiting, the predicted number of people entering and / or the predicted number of people leaving corresponding to at least one user type and the target time period;

[0206] Based on the predicted number of people entering and / or predicted number of people leaving, the target number of people retained in the target building corresponding to the target time period is determined, and the target number of people retained is used to predict the power load of the target building.

[0207] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0208] Cluster the predicted number of people entering and exiting according to each user type to obtain the target predicted number of people entering and exiting corresponding to each user type;

[0209] Determine the predicted number of people entering and exiting according to the predicted number of people entering and exiting each target and the entry probability function corresponding to each user type;

[0210] According to the predicted number of people entering and leaving each target and the survival probability function corresponding to each user type, the predicted number of people leaving is determined.

[0211] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0212] For each user type, the product of the predicted number of in-and-out persons corresponding to the user type and the entry probability function corresponding to the user type is determined as the predicted number of in-and-out persons.

[0213] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0214] For each user type, the product of the predicted number of people entering and leaving the user type and the survival probability function corresponding to the user type is determined as the predicted number of people leaving the user type.

[0215] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0216] For a user type, the difference between the predicted number of people entering and the predicted number of people leaving is determined as the initial number of people retained corresponding to the user type;

[0217] Determine the target retention number based on the initial retention number.

[0218] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0219] Collect historical user monitoring data of the target building within a historical time period, the historical user monitoring data including the historical number of people entering, historical entry time, historical number of people leaving, and historical leaving time corresponding to each user type and historical time period;

[0220] According to the historical user monitoring data, the initial number prediction model is iteratively trained to obtain the target number prediction model.

[0221] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0222] For each user type, obtain the entry time of each user in the user type into the target building and the exit time of each user from the target building;

[0223] According to the entry time, determine the number of people entering corresponding to multiple preset moments and the total number of people entering within the preset time period, each preset moment being a moment within the preset time period;

[0224] According to the departure time, determine the number of people leaving corresponding to multiple preset moments and the total number of people leaving within the preset time period;

[0225] The entry probability function is obtained according to the ratio of each number of people entering and the total number of people entering, and the survival probability function is obtained according to the ratio of each number of people leaving and the total number of people leaving.

[0226] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0227] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0228] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for predicting power load, characterized in that: The method comprises: Determine the predicted number of people entering and leaving the target building corresponding to the target time period according to the target number prediction model corresponding to the target building; Determine, based on the predicted number of people entering and exiting, a predicted number of people entering and / or a predicted number of people leaving corresponding to the target time period for at least one user type; The target number of people to be retained in the target building corresponding to the target time period is determined based on the predicted number of people entering and / or the predicted number of people to be left, and the target number of people to be retained is used to predict the power load of the target building.

2. The method according to claim 1, characterized in that Determining the predicted number of people entering and leaving the target time period according to the predicted number of people entering and leaving the target time period for at least one user type includes: Clustering the predicted number of people entering and exiting according to each user type to obtain the target predicted number of people entering and exiting corresponding to each user type; Determining the predicted number of people entering and exiting according to each of the target predicted number of people entering and exiting and the entry probability function corresponding to each of the user types; The predicted number of people leaving is determined according to the target predicted number of people entering and leaving and the survival probability function corresponding to each user type.

3. The method according to claim 2, characterized in that The step of determining the predicted number of people entering and exiting the target according to the predicted number of people entering and exiting the target and the entry probability function corresponding to each user type comprises: For each of the user types, the product of the predicted number of people entering and exiting corresponding to the user type and the entry probability function corresponding to the user type is determined as the predicted number of people entering.

4. The method according to claim 2, characterized in that: The step of determining the predicted number of people leaving the site according to the predicted number of people entering and leaving the site and the survival probability function corresponding to each of the user types includes: For each of the user types, the product of the predicted number of people entering and leaving the user type and the survival probability function corresponding to the user type is determined as the predicted number of people leaving the user type.

5. The method according to any one of claims 1 to 4, characterized in that: The determining, according to the predicted number of people entering and / or the predicted number of people leaving, a target number of people retained in the target building corresponding to the target time period includes: For a user type, the difference between the predicted number of people entering and the predicted number of people leaving is determined as the initial number of people retained corresponding to the user type; The target number of retained persons is determined based on the initial number of retained persons.

6. The method according to claim 1, characterized in that The method further comprises: Collecting historical user monitoring data of the target building within a historical time period, the historical user monitoring data including the historical number of people entering, the historical entry time, the historical number of people leaving, and the historical leaving time corresponding to each user type and the historical time period; The initial number of people prediction model is iteratively trained according to the historical user monitoring data to obtain the target number of people prediction model.

7. The method according to claim 2, characterized in that The method further comprises: For each of the user types, obtaining the entry time of each user in the user type entering the target building and the exit time of each user leaving the target building; According to the entry time, determining the number of people entering corresponding to a plurality of preset moments and the total number of people entering within a preset time period, each of the preset moments being a moment within the preset time period; According to the departure time, determine the number of people leaving corresponding to the plurality of preset moments and the total number of people leaving within the preset time period; The entry probability function is obtained according to the ratio of each of the entering numbers to the total entering number, and the survival probability function is obtained according to the ratio of each of the leaving numbers to the total leaving number.

8. A power load forecasting device, characterized in that: The device comprises: A first determination module is used to determine the predicted number of people entering and exiting the target building corresponding to the target time period according to a target number prediction model corresponding to the target building; A second determination module is used to determine the predicted number of people entering and leaving the target time period, based on the predicted number of people entering and leaving the target time period, for at least one user type; The third determination module is used to determine the target number of people to be retained in the target building corresponding to the target time period according to the predicted number of people entering and / or the predicted number of people to be left, and the target number of people to be retained is used to predict the power load of the target building.

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

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

11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images