Methods, devices, equipment and storage media for identifying electrical loads

Abstract

This application discloses a method, apparatus, device, and storage medium for identifying electrical load, relating to the field of power system technology, to at least solve the technical problem of low load identification results when identifying production line loads in complex operating scenarios in related technologies. The method includes: constructing multiple initial models corresponding to preset time periods; optimizing the model parameters of each initial model based on historical load data of each production line in each preset time period to obtain target models corresponding to multiple preset time periods; inputting observed load data before the current time into the target model corresponding to the preset time period to which the preset time period before the current time belongs, to obtain the target operating states of each production line under the current total power and the target energy distribution of each production line under each target operating state.

Description

Technical Field

[0001] This application relates to the field of power system technology, and in particular to a method, apparatus, equipment and storage medium for identifying electrical loads. Background Technology

[0002] With the development of the social economy, industrial electricity consumption has increased year by year, and the imbalance between power generation and demand has become increasingly severe. Although the widespread integration of distributed photovoltaic power has provided backup support for the load gap, accurate and efficient orderly power consumption and demand response remain important means to ensure the safe and stable operation of the power grid. Furthermore, the refined perception of the sub-loads of industrial users can support more accurate load forecasting and demand response planning.

[0003] Most relevant load identification technologies focus on event detection and feature analysis in traditional signal analysis methods, such as industrial load identification methods based on random forests and steady-state waveforms, and industrial event identification methods based on the mapping relationship between interpretation space and feature space. These feature-based load identification methods rely on correctly detecting the "switching" events of equipment in the production line. They are suitable for industrial scenarios where the production line equipment is stable and the total number of devices is small, enabling event detection and type identification through feature increment thresholds. However, when identifying the load of a production line involving a large number of devices and frequent equipment state switching, the load identification results are low due to the limitation of the identification dimensions of these relevant load identification methods. Summary of the Invention

[0004] This invention provides a method, apparatus, device, and storage medium for identifying electrical loads, to at least solve the technical problem of low load identification results when identifying production line loads in complex operating scenarios in related technologies. The technical solution of this invention is as follows:

[0005] According to a first aspect of the present invention, an electrical load identification method is provided. The method includes: constructing multiple initial models corresponding to preset time periods; the initial models characterize the correlation between the total power of each production line and the various operating states of each production line and the energy distribution of the corresponding production line within each preset time period; optimizing the model parameters of each initial model based on the historical load data of each production line in each preset time period, with the objective of maximizing the probability of each historical operating state of each production line under the historical total power of each production line, to obtain target models corresponding to the multiple preset time periods; the historical load data includes a historical active power sequence composed of multiple historical active power of each production line in each preset time period, and the number of historical operating states and the number of historical active states under each historical active power. The historical distribution probability of each historical operating state and the historical transition probability of each historical operating state switching for each production line in all production lines are obtained; the observed load data are obtained from the observed operating data of each production line within a preset time period before the current time, and the observed load data are input into the target model corresponding to the preset time period to which the preset time period before the current time belongs, to obtain the target operating states of each production line under the current total power and the target energy distribution of each production line under each target operating state; wherein, each target operating state is obtained with the goal of maximizing the probability of each operating state to which each production line belongs under the observed total power of each production line; the target energy distribution of each production line under each target operating state is obtained with the goal of maximizing the active power probability to which each production line belongs under each target operating state of each production line.

[0006] The above-mentioned multiple preset time periods constitute a time cycle.

[0007] Optionally, multiple preset time periods are consecutive, meaning that multiple consecutive preset time periods form a time cycle.

[0008] Multiple preset time periods can be divided according to the correlation or similarity of the operating status of each production line within the time range being higher than the preset similarity, so that the correlation or similarity of the operating status of each production line in the same preset time period of different time cycles is relatively high, so as to ensure that the operating rules of the production line represented by each target model are more accurate, thereby avoiding the influence of time factors.

[0009] In one possible implementation, the target model includes a Gaussian mixture model and a hidden Markov model. Current load data is obtained from the current observed operating data at the current time under the current total power of each production line. This includes: calling the Gaussian mixture model in the target model to cluster the current operating observation data of each production line, obtaining the number of multiple observed operating states included in each production line and the distribution probability of each production line being in each observed operating state under each observed active power; the multiple observed active powers of each production line constitute an observed active power sequence; using the hidden Markov model in the target model, based on the switching frequency of operating state switching between different time-adjacent observed operating states in each production line, the transition probability of each observed operating state switching of each production line in all production lines is determined.

[0010] The above-mentioned clustering of the current operation observation data for each production line specifically includes extracting and clustering the current operation observation data for each production line according to the equipment characteristics and / or power consumption characteristics and / or load identifiers of each production line.

[0011] Optionally, given that the hierarchical hidden Markov model can significantly improve the application effect and interpretability of the model in complex data analysis and prediction by introducing multi-level abstraction and representation capabilities, and can better handle multi-scale, multi-dimensional, and multi-production line operation observation data, the hidden Markov model can be set as a hierarchical hidden Markov model to more determine the probability of operation state switching from multiple dimensions of production state and switching state.

[0012] Among the above implementation methods, the Gaussian mixture distribution model is highly flexible and has strong expressive power. It performs well in handling complex data distributions and various applications, and can more accurately identify the status of the production line in complex working conditions such as a large number of production line devices and frequent device status switching.

[0013] In another possible implementation, the initial model includes a Gaussian mixture model and a hidden Markov model. Before optimizing the model parameters of each initial model based on the historical load data of each production line in each preset time period, with the objective of maximizing the probability of each production line belonging to each historical operating state under the historical total power of each production line, the method further includes: acquiring historical operating observation data of each production line in each preset time period; using the Gaussian mixture model in the initial model to cluster the historical operating observation data of each production line to obtain the number of historical operating states included in each production line and the distribution probability of each production line being in each historical operating state under each observed historical active power; the multiple historical active powers of each production line constitute a historical active power sequence.

[0014] Using the Hidden Markov Model in the initial model, the transition probability of each historical operating state of each production line is determined based on the switching frequency of the operating state switching between different historical operating states that are adjacent in time in each production line.

[0015] Another possible implementation, the Gaussian mixture distribution model specifically includes:

[0016]

[0017] Among them, B i x represents any one of the 1 to m observed running states. i The probability distribution of different states corresponding to the observed active power; Let the observed power of production line i be at time t. At that time, observe the operating status as follows: The probability of μ; i,s and σ i,s Production line i is in the observation and operation status. The average power and standard deviation over time.

[0018] Another possible implementation of the Hidden Markov Model specifically includes:

[0019]

[0020] in, This represents the probability of production line i switching its operating state within the h-th hour; For production line i from time t-1 to time t, the operating state x t-1 i Transition to running state The probability of n; i,h (s1,s2) For production line i in the corresponding observed active power sequence y i The running status before and after the middle of the time is from x t-1 i Transfer to x t-1 i The number, N, is the number of production lines i in the observed power sequence y. i The total number of observed operational states included above.

[0021] In another possible implementation, the initial model and the target model specifically include:

[0022]

[0023] in, This indicates that the i-th production line starts from x in the h-th hour.t-1 i Running status The transition probability of the running state, When production line i is in operating state at time t: The total observed power is y t The probability of.

[0024] In another possible implementation, the initial model and the target model specifically include:

[0025]

[0026] in, At time t, production line i is in the following operating state: The observed active power is The probability of.

[0027] According to a second aspect of the present invention, an electrical load identification device is provided. The device includes: a construction unit, configured to construct multiple initial models corresponding to preset time periods; the initial models characterize the correlation between the total power of each production line and the various operating states of each production line and the energy distribution of each production line within each preset time period; and a training unit, configured to optimize the model parameters of each initial model based on historical load data of each production line within each preset time period, with the objective of maximizing the probability of each historical operating state of each production line under the historical total power of each production line, to obtain target models corresponding to multiple preset time periods; the historical load data includes a historical active power sequence composed of multiple historical active power of each production line within each preset time period, and includes the number of historical operating states and the number of historical active power under each historical active power. The system includes: historical distribution probabilities of each historical operating state under various power levels and historical transition probabilities of each historical operating state switching for each production line in all production lines; an identification unit, used to obtain observed load data from the observed operating data of each production line within a preset time period before the current time, and input the observed load data into the target model corresponding to the preset time period before the current time, to obtain the target operating states of each production line under the current total power and the target energy distribution of each production line under each target operating state; wherein, each target operating state is obtained with the goal of maximizing the probability of each operating state belonging to each production line under the observed total power of each production line; the target energy distribution of each production line under each target operating state is obtained with the goal of maximizing the active power probability belonging to each production line under each target operating state of each production line.

[0028] According to a third aspect of the present invention, an electrical load identification system is provided, configured to perform an electrical load identification method as described in the first aspect and any possible implementation thereof.

[0029] According to a fourth aspect of the present invention, an electrical device is provided, comprising: a processor and a memory for storing processor-executable instructions; wherein the processor is configured to execute the executable instructions to implement an electrical load identification method as described in the first aspect and any possible implementation thereof.

[0030] According to a fifth aspect of the present invention, a computer-readable storage medium is provided, on which instructions are stored, such that when the instructions in the computer-readable storage medium are executed by a processor of an electrical device, the electrical device is enabled to perform an electrical load identification method as described in the first aspect and any possible implementation thereof.

[0031] According to a sixth aspect of the embodiments of this application, a computer program product is provided, the computer program product including computer instructions, which, when executed on an electrical device, cause the electrical device to perform the power load identification method of the first aspect and any possible implementation thereof.

[0032] The technical solution provided by the embodiments of the present invention brings at least the following beneficial effects: First, the correlation and operating rules between the total power of the production line and the various operating states of each production line and the corresponding energy distribution of each production line are constructed to obtain an initial model. Then, with the goal of maximizing the probability of each historical operating state of each production line under the historical total power of each production line, the model parameters of the initial model are optimized and trained based on historical load data of multiple different preset time periods to obtain target models that match different preset time periods. At the same time, the target models are obtained by optimizing the model parameters with the goal of maximizing the probability of each historical operating state, so that the total power of the production line and the various operating states of each production line represented by the target model are the result of the simultaneous occurrence and maximum matching probability of the parameters of both, thereby ensuring the accuracy of the target model in identifying the operating state. Using the target model that matches the preset time period that matches the current time (i.e., the target model corresponding to the preset time period before and adjacent to the current time), the target operating state of each production line under the current total power and the target energy distribution of each production line under each target operating state are determined to intelligently realize the identification of the operating state and load power consumption of each production line.

[0033] Using the aforementioned power load identification method, and based on the target model, the number of operating states under each observed active power, the probability distribution of each operating state, and the transition probability of each production line switching operating states, the target operating states of each production line can be accurately obtained, with the goal of maximizing the probability of each operating state belonging to each production line under the observed total power of each production line. This ensures accurate identification of the operating states of each production line under different complexity scenarios. At the same time, it can also accurately obtain the target energy distribution of each production line under each target operating state, with the goal of maximizing the probability of active power belonging to each production line under each target operating state. This ensures accurate identification of the power load of each production line under different complexity scenarios.

[0034] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0035] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application, and do not constitute an undue limitation of this application.

[0036] Figure 1 This is a flowchart illustrating an electrical load identification method according to an exemplary embodiment;

[0037] Figure 2 This is a schematic block diagram illustrating an electrical load identification device according to an exemplary embodiment;

[0038] Figure 3 This is a schematic diagram of an electrical device according to an exemplary embodiment. Detailed Implementation

[0039] To enable those skilled in the art to better understand the technical solutions of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0040] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application 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 this application described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0041] Before providing a detailed description of the power load identification method provided in the embodiments of this application, let's briefly introduce the application scenarios involved in the embodiments of this application.

[0042] With the development of the social economy, industrial electricity consumption has increased year by year, and the imbalance between power generation and demand has become increasingly severe. Although the widespread integration of distributed photovoltaic power has provided backup support for load gaps, accurate and efficient orderly power consumption and demand response remain important means to ensure the safe and stable operation of the power grid. Refined perception of individual loads of industrial users can support more accurate load forecasting and demand response planning. Non-intrusive load identification technology can obtain information on the usage of various user equipment at low cost, enabling online monitoring and analysis of power load, which is of great significance for supporting applications such as load forecasting and demand response.

[0043] Load identification technology is applied to residential electricity consumption analysis. Benefiting from the concentrated types of residential electrical equipment, similar usage patterns, and easy data collection, load decomposition strategies are increasingly focusing on representation learning methods under various deep neural network models. Simultaneously, to increase the generalization and ease of use of such supervised learning algorithms, emerging algorithms such as transfer learning and reinforcement learning have become major research directions in residential load identification. However, industrial load identification applications involve diverse user types, a wide range of equipment types, and complex equipment operating patterns, significantly differing from residential load identification scenarios. This makes it difficult to implement large-scale data-driven network learning methods. Most related methods focus on event detection and feature analysis in traditional signal analysis methods: for example, industrial load identification methods based on random forests and steady-state waveforms; and industrial event identification methods based on the mapping relationship between interpretation space and feature space. The aforementioned feature-based methods rely on correctly detecting the "switching" events of equipment. In general, they can achieve event detection and type identification through feature increment thresholds in industrial scenarios where equipment operation is stable and the total number of equipment is small. However, for production line type identification scenarios involving a large number of electrical appliances, frequent equipment switching, and small load magnitude, the algorithm needs to be deeply customized to adapt to various complex working conditions, which has certain limitations.

[0044] Research has found that in residential load identification applications, Hidden Markov Models (HMMs) were the most widely used analytical method before the popularization of deep network identification algorithms. By using the expectation-maximization algorithm to find the optimal hidden state sequence, they achieve the identification and decomposition of electrical loads. HMMs possess the ability to handle the simultaneous occurrence of complex switching events and have relatively low data acquisition requirements, making them a preferred method for identifying complex industrial loads. A single production line contains a complex number of devices and complex production processes. Using a single Gaussian distribution to describe the load can introduce errors when performing probability-driven state judgments in scenarios involving multiple overlapping production lines. Furthermore, it also affects energy decomposition. Therefore, a possible approach is to cluster the load of a single production line based on a Gaussian mixture distribution before training the model to distinguish between different operating states of the same production line.

[0045] Therefore, this application provides a method for identifying electricity load. First, it constructs an initial model by establishing the correlation and operational patterns between the total power of the production line and the various operating states and corresponding energy distributions of each production line. Then, with the objective of maximizing the probability of each production line's historical operating state under its historical total power, the model parameters of the initial model are optimized and trained based on historical load data from multiple preset time periods. This yields target models that match different preset time periods. These target models are obtained by optimizing model parameters with the objective of maximizing the probability of each historical operating state, ensuring that the target model represents the situation where the total power of the production line and the various operating states of each production line occur simultaneously and have the highest probability of matching. This guarantees the accuracy of the target model in identifying operating states. Finally, using the target model that matches a preset time period (i.e., the target model corresponding to a preset time period before and adjacent to the current time), the target operating states of each production line under the current total power and the target energy distributions of each production line under each target operating state are determined, thereby intelligently identifying the operating states and load electricity consumption of each production line.

[0046] The power load identification method provided in this application can be applied to power equipment used for power load identification. For ease of understanding, the power load identification method provided in this application will be described in detail below with reference to the accompanying drawings.

[0047] Figure 1 This is a flowchart illustrating an electrical load identification method according to an exemplary embodiment, such as... Figure 1 As shown, the electrical load identification method includes the following steps.

[0048] S11, construct each initial model corresponding to multiple preset time periods.

[0049] The initial model represents the correlation between the total power of each production line and the various operating states of each production line and the corresponding energy distribution of each production line within each preset time period.

[0050] The above-mentioned multiple preset time periods constitute a time cycle.

[0051] Optionally, multiple preset time periods are consecutive, meaning that multiple consecutive preset time periods form a time cycle.

[0052] Multiple preset time periods can be divided according to the correlation or similarity of the operating status of each production line within the time range being higher than the preset similarity, so that the correlation or similarity of the operating status of each production line in the same preset time period of different time cycles is relatively high, so as to ensure that the operating rules of the production line represented by each target model are more accurate, thereby avoiding the influence of time factors.

[0053] The operating status of the aforementioned production line can be understood as the start-up and shutdown status of each piece of equipment on the production line.

[0054] S12, with the goal of maximizing the probability of each production line's historical operating state under the historical total power of each production line, optimizes the model parameters of each initial model based on the historical load data of each production line in each preset time period, and obtains target models corresponding to multiple preset time periods.

[0055] Historical load data includes a sequence of historical active power for each production line in each preset time period, the number of historical operating states under each historical active power, the historical distribution probability of each historical operating state under each historical active power, and the historical transition probability of each historical operating state switching for each production line in all production lines.

[0056] In one implementation, the initial model includes a Gaussian mixture model (also known as a Gaussian Mixture Model, GMM) and a Hidden Markov Model. The method, which aims to maximize the probability of each production line belonging to each historical operating state under the historical total power of each production line, optimizes the model parameters of each initial model based on the historical load data of each production line in each preset time period to obtain the target model corresponding to each preset time period. The method further includes: acquiring historical operating observation data of each production line in each preset time period; using the Gaussian mixture distribution model in the initial model to cluster the historical operating observation data of each production line to obtain the number of historical operating states included in each production line and the distribution probability of each production line being in each historical operating state under each observed historical active power; the multiple historical active powers of each production line constitute a historical active power sequence; using the Hidden Markov Model in the initial model, based on the switching frequency of operating state switching between different historical operating states that are temporally adjacent in each production line, determining the transition probability of each historical operating state switching for each production line in all production lines.

[0057] S13. Obtain the observed load data from the observed operating data of each production line within the preset time period before the current time, and input the observed load data into the target model corresponding to the preset time period to which the preset time period before the current time belongs, so as to obtain the target operating state of each production line under the current total power and the target energy distribution of each production line under each target operating state.

[0058] The target energy distribution of each of the above production lines is the target power corresponding to each production line.

[0059] Among them, each target operating state is obtained by maximizing the probability of each operating state belonging to each production line under the observed total power of each production line; the target energy distribution of each production line under each target operating state is obtained by maximizing the probability of active power belonging to each production line under each target operating state of each production line.

[0060] In one implementation, the target model includes a Gaussian mixture model and a hidden Markov model. The current load data is obtained from the current observed operating data at the current time under the current total power of each production line. This includes: calling the Gaussian mixture model in the target model to cluster the current operating observation data of each production line, obtaining the number of multiple observed operating states included in each production line and the distribution probability of each production line being in each observed operating state under each observed active power; the multiple observed active powers of each production line constitute an observed active power sequence; using the hidden Markov model in the target model, based on the switching frequency of operating state switching between different time-adjacent observed operating states in each production line, the transition probability of each observed operating state switching of each production line in all production lines is determined.

[0061] The above-mentioned clustering of the current operation observation data for each production line specifically includes extracting and clustering the current operation observation data for each production line according to the equipment characteristics and / or power consumption characteristics and / or load identifiers of each production line.

[0062] Optionally, given that the hierarchical hidden Markov model can significantly improve the application effect and interpretability of the model in complex data analysis and prediction by introducing multi-level abstraction and representation capabilities, and can better handle multi-scale, multi-dimensional, and multi-production line operation observation data, the hidden Markov model can be set as a hierarchical hidden Markov model (FHMM model) to more definitively determine the probability of operation state switching from multiple dimensions of production state and switching state.

[0063] In the above embodiments, given that the Gaussian mixture distribution model is highly flexible and has strong expressive power, it performs well in handling complex data distributions and various applications, and can more accurately identify the status of the production line in complex working conditions such as a large number of production line devices and frequent device status switching.

[0064] Optionally, for continuously varying active power, a Gaussian distribution is used to describe the probability distribution (i.e., distribution probability) of the observations in the operating and stopped states. The mixed Gaussian distribution model in the target model and the initial model mentioned above specifically includes the model represented by the following formula 1.

[0065]

[0066] Among them, B i x represents any one of the 1 to m observed running states. i The probability distribution of different states corresponding to the observed active power; Let the observed power of production line i be at time t. At that time, observe the operating status as follows: The probability of μ; i,s and σ i,s Production line i is in the observation and operation status. The average power and standard deviation over time.

[0067] Optionally, the hidden Markov models in the target model and the initial model mentioned above specifically include the model represented by the following formula 2.

[0068]

[0069] in, This represents the probability of production line i switching its operating state within the h-th hour; For production line i from time t-1 to time t, the operating state x t-1 i Transition to running state The probability of n; i,h (s1,s2) For production line i in the corresponding observed active power sequence y i The running status before and after the middle of the time is from x t-1 i Transfer to x t-1 i The number, N, is the number of production lines i in the observed power sequence y. i The total number of observed operational states included above.

[0070] Optionally, based on the determination of the initial model and related model parameters in the above embodiments, the hidden operating state (hereinafter referred to as operating state) of the production line corresponding to the total power curve can be further analyzed. Since the current state in the Markov chain model is only related to the previous state, the problem of determining the operating state of the production line can be described as follows: In the scenario where the operating state of the equipment is known at time t-1, the total power at time t is y. t The problem involves solving the optimal state sequence of each electrical appliance at time t, and based on the probabilities of each production line under various operating states at time t, the operating state corresponding to the maximum probability is taken as the current state (i.e., the target state). Therefore, the above target model and initial model also include the model represented by the following (Equation 3).

[0071]

[0072] in, This indicates that the i-th production line starts from x in the h-th hour. t-1 i Running status The transition probability of the running state, When production line i is in operating state at time t: The total observed power is y t The probability of.

[0073] Optionally, the initial model and the target model mentioned above also include the model represented by Equation 4 below.

[0074]

[0075] in, At time t, production line i is in the following operating state: The observed active power is The probability of.

[0076] Optionally, due to the complexity of the solution, in practical applications, the average power distribution of each production line under its current operating state (i.e., the target operating state) is generally used as the target power of the production line under the target operating state at the current moment. Further, the average Gaussian distribution of active power under the current operating state (i.e., the target operating state) of each production line can be chosen as the target power of the production line under the target operating state at the current moment.

[0077] In some embodiments, clustering is performed using active power and reactive power as examples of observed operating parameters. The probability density function of the Gaussian distribution is shown in the following formula (5).

[0078]

[0079] Where K is the number of operating states of a single production line, i.e., the number of clusters; α k Let be the prior probability that a data point belongs to the k-th Gaussian. The initial value only needs to satisfy all α k The sum is 1; p(x,y|μ 1,k ,σ 1,k ,μ 2,k ,σ 2,k ) represents the probability density of the k-th Gaussian distribution; μ 1,k σ 1,k Let μ be the mean and variance of the active power in the k-th operating state of a single production line; 2,k σ 2,k Let K be the mean and variance of reactive power in the k-th operating state of a single production line. During cluster analysis, K is a given value, and other parameters are estimated using the expectation-maximum (EM) algorithm. Since the actual production line typically cannot provide a specific number of operating states, a cluster evaluation algorithm is used. The sum of squared errors between clusters is considered a function of the number of clusters K. The optimal value of K is located at the "inflection point" of the curve, and clustering is completed based on a Gaussian mixture distribution model.

[0080] The construction process of the FHMM model based on GMM clustering is basically the same as the three steps mentioned above. The probability distribution of the observations of the production line equipment under the start-up and shutdown states in formula (1) is transformed into the probability distribution of the observations under multiple different states. When performing load decomposition, the states contained in a single equipment in formulas (2) and (3) can be synchronously added to achieve FHMM load decomposition of a single production line under multiple states. Finally, by adding the pre-statistical prior information of the production line operation rules to the FHMM model and setting different state transition probabilities in different time sections, the accuracy of actual load decomposition can be improved.

[0081] Through the above implementation method, the correlation and operating rules between the total power of the production line and the various operating states of each production line and the corresponding energy distribution of each production line are first constructed to obtain an initial model. Then, with the goal of maximizing the probability of each historical operating state of each production line under the historical total power of each production line, the model parameters of the initial model are optimized and trained based on historical load data of multiple different preset time periods to obtain target models that match different preset time periods. At the same time, the target models are obtained by optimizing the model parameters with the goal of maximizing the probability of each historical operating state, so that the total power of the production line and the various operating states of each production line represented by the target model are the result of the simultaneous occurrence and maximum matching probability of the parameters of both, thereby ensuring the accuracy of the target model in identifying the operating state. Using the target model that matches the preset time period that matches the current time (i.e., the target model corresponding to the preset time period before and close to the current time), the target operating state of each production line under the current total power and the target energy distribution of each production line under each target operating state are determined to intelligently realize the identification of the operating state and load power consumption of each production line.

[0082] Therefore, according to the target model, based on the number of operating states under each observed active power, the probability distribution of each operating state, and the transition probability of each production line switching operating states, the target operating states of each production line can be accurately obtained with the goal of maximizing the probability of each operating state belonging to each production line under the observed total power of each production line. This ensures accurate identification of the operating states of each production line under different complexity scenarios. At the same time, the target energy distribution of each production line under each target operating state can also be accurately obtained with the goal of maximizing the probability of active power belonging to each production line under each target operating state. This ensures accurate identification of the power load of each production line under different complexity scenarios.

[0083] To achieve the above functions, the electrical load identification device includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art will readily recognize that, based on the algorithmic steps of the examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0084] This application embodiment also provides a method such as Figure 2 The power load identification device shown includes a construction unit 201, a training unit 202, and an identification unit 203.

[0085] The construction unit 201 is used to construct each initial model corresponding to multiple preset time periods. The initial model represents the correlation between the total power of each production line and the various operating states of each production line and the energy distribution of each production line within each preset time period.

[0086] Training unit 202 is used to optimize the model parameters of each initial model based on the historical load data of each production line in each preset time period, with the goal of maximizing the probability of each historical operating state of each production line under the historical total power of each production line. The historical load data includes the historical active power sequence composed of multiple historical active power of each production line in each preset time period, the number of historical operating states under each historical active power, the historical distribution probability of each historical operating state under each historical active power, and the historical transition probability of each historical operating state of each production line in all production lines switching operating states.

[0087] The identification unit 203 is used to obtain observed load data from the observed operating data of each production line within a preset time period before the current time, and input the observed load data into the target model corresponding to the preset time period to which the preset time period before the current time belongs, to obtain the target operating states of each production line under the current total power and the target energy distribution of each production line under each target operating state; wherein, each target operating state is obtained with the goal of maximizing the probability of each operating state to which each production line belongs under the observed total power of each production line; the target energy distribution of each production line under each target operating state is obtained with the goal of maximizing the probability of active power to which each production line belongs under each target operating state of each production line.

[0088] In one possible implementation, the target model includes a Gaussian mixture model and a hidden Markov model; the identification unit 203 is specifically used to: call the Gaussian mixture model in the target model to cluster the current operation observation data of each production line, thereby obtaining the number of multiple observed operation states included in each production line and the distribution probability of each production line being in each observed operation state under each observed active power; the multiple observed active powers of each production line constitute an observed active power sequence; using the hidden Markov model in the target model, based on the switching frequency of operation state switching between different time-adjacent observed operation states in each production line, determine the transition probability of each observed operation state switching of each production line in all production lines.

[0089] In another possible implementation, the initial model includes a Gaussian mixture model and a hidden Markov model; the training unit 202 is further used to: acquire historical operation observation data of each production line in each preset time period; use the Gaussian mixture model in the initial model to cluster the historical operation observation data of each production line to obtain the number of multiple historical operation states included in each production line and the distribution probability of each production line in each historical operation state under each observed historical active power; the multiple historical active powers of each production line constitute a historical active power sequence; use the hidden Markov model in the initial model to determine the transition probability of each historical operation state of each production line in all production lines to switch operation states based on the switching frequency of operation state switching between different historical operation states that are adjacent in time in each production line.

[0090] Another possible implementation, the Gaussian mixture distribution model specifically includes:

[0091]

[0092] Among them, B i x represents any one of the 1 to m observed running states. i The probability distribution of different states corresponding to the observed active power; Let the observed power of production line i be at time t. At that time, observe the operating status as follows: The probability of μ; i,s and σ i,s Production line i is in the observation and operation status. The average power and standard deviation over time.

[0093] Another possible implementation of the Hidden Markov Model specifically includes:

[0094]

[0095] in, This represents the probability of production line i switching its operating state within the h-th hour; For production line i from time t-1 to time t, the operating state x t-1 i Transition to running state The probability of n; i,h (s1,s2) For production line i in the corresponding observed active power sequence y i The running status before and after the middle of the time is from x t-1 i Transfer to x t-1 i The number, N, is the number of production lines i in the observed power sequence y. i The total number of observed operational states included above.

[0096] In another possible implementation, the initial model and the target model specifically further include:

[0097]

[0098] in, This indicates that the i-th production line starts from x in the h-th hour. t-1 i Running status to x t i The transition probability of the running state, When production line i is in operating state at time t: The total observed power is y t The probability of.

[0099] In another possible implementation, the initial model and the target model specifically further include:

[0100]

[0101] in, At time t, production line i is in the following operating state: The observed active power is The probability of.

[0102] Regarding the apparatus in the above embodiments, the specific manner in which each unit module performs its operations has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0103] Figure 3 This is a schematic diagram of an electrical device provided in this application. Figure 3The power device 60 may include at least one processor 601 and a memory 603 for storing processor-executable instructions. The processor 601 is configured to execute the instructions in the memory 603 to implement the power load identification method in the following embodiments.

[0104] In addition, the power equipment 60 may also include a communication bus 602, at least one communication interface 604, an input device 606, and an output device 605.

[0105] The processor 601 may be a processor (central processing unit, CPU), a microprocessor unit, an ASIC, or one or more integrated circuits for controlling the execution of the program of the present application.

[0106] The communication bus 602 may include a path for transmitting information between the aforementioned components.

[0107] Communication interface 604 uses any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0108] Input device 606 is used to receive input signals and output device 605 is used to output signals.

[0109] The memory 603 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. The memory may exist independently and be connected to the processing unit via a bus. The memory may also be integrated with the processing unit.

[0110] The memory 603 stores instructions for executing the scheme of this application, and the processor 601 controls the execution. The processor 601 executes the instructions stored in the memory 603 to realize the functions of the method of this application.

[0111] In a specific implementation, as one embodiment, the processor 601 may include one or more CPUs, for example... Figure 3 CPU0 and CPU1 in the CPU.

[0112] In a specific implementation, as one example, the power device 60 may include multiple processors, such as... Figure 3 Processors 601 and 607 are described herein. Each of these processors may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. A processor here may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0113] The power equipment is like Figure 3 The diagram includes a processor 601 and a memory 603 for storing executable instructions of the processor 601; wherein the processor 601 is configured to execute executable instructions to implement the power load identification method as described in any of the possible embodiments above. And it can achieve the same technical effect, so to avoid repetition, it will not be described again here.

[0114] This application also provides a computer-readable storage medium, which, when the instructions in the computer-readable storage medium are executed by the processor of a control device or control apparatus, enables the control device or control apparatus to perform the power load identification method as described in any of the possible embodiments above. And it can achieve the same technical effect; to avoid repetition, it will not be described again here.

[0115] This application also provides a computer program product, including a computer program or instructions, which are executed by a processor as described in any of the possible implementations of the power load identification method above. Since the same technical effects can be achieved, further details are omitted here to avoid repetition.

[0116] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0117] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for identifying electrical loads, characterized in that, The method includes: Construct initial models corresponding to multiple preset time periods; each initial model represents the correlation between the total power of each production line and the operating states of each production line and the energy distribution of each production line within each preset time period. With the objective of maximizing the probability of each production line's historical operating state under its historical total power, the model parameters of each initial model are optimized based on the historical load data of each production line in each preset time period to obtain the target models corresponding to the multiple preset time periods. The historical load data includes a historical active power sequence composed of multiple historical active power of each production line in each preset time period, as well as the number of historical operating states under each historical active power, the historical distribution probability of each historical operating state under each historical active power, and the historical transition probability of each historical operating state switching of each production line in all production lines. The observed load data is obtained from the observed operating data of each production line within a preset time period that has the smallest time interval before the current time and with the current time. This observed load data is then input into the target model corresponding to the preset time period that has the smallest time interval before the current time. This yields the target operating states of each production line under the current total power and the target energy distribution of each production line under each target operating state. The target operating states are obtained by maximizing the probability of each operating state belonging to each production line under the observed total power. The target energy distribution of each production line under each target operating state is obtained by maximizing the probability of active power belonging to each production line under each target operating state. The target model includes a Gaussian mixture model and a hidden Markov model. Obtaining the current load data from the current observed operating data of each production line at the current time under the current total power includes: The Gaussian mixture distribution model in the target model is invoked to cluster the current observed operating data of each production line, thereby obtaining the number of multiple observed operating states included in each production line and the distribution probability of each production line being in each observed operating state under each observed active power; the multiple observed active powers of each production line constitute an observed active power sequence. Using the Hidden Markov Model in the target model, the transition probability of each observed operating state switching in all production lines is determined based on the switching frequency of different time-adjacent observed operating states in each production line.

2. The electrical load identification method according to claim 1, characterized in that, The initial model includes a Gaussian mixture model and a hidden Markov model; before optimizing the model parameters of each initial model based on the historical load data of each production line in each preset time period, with the objective of maximizing the probability of each production line's historical operating state under the historical total power of each production line, to obtain the target model corresponding to each of the multiple preset time periods, the method further includes: Obtain historical operational observation data for each production line during each preset time period; Using the Gaussian mixture distribution model in the initial model, the historical operation observation data of each production line are clustered to obtain the number of historical operation states included in each production line and the distribution probability of each production line in each historical operation state under each observed historical active power; the multiple historical active powers of each production line constitute a historical active power sequence. Using the Hidden Markov Model in the initial model, the transition probability of each historical operating state of each production line in all production lines is determined based on the switching frequency of the operating state switching between different historical operating states that are adjacent in time in each production line.

3. The method for identifying electrical loads according to claim 1 or 2, characterized in that, The Gaussian mixture distribution model specifically includes: in, B i Indicates 1 to m Any observation operation state among the observation operation states x i The probability distribution of different states corresponding to the observed active power; P( y t i |x t i ) represents the production line at time t. i At the observed power y t i At that time, the observed operating status was x t i The probability of; μ i,s and σ i,s They are production lines i In observation operation status x t i The average power and standard deviation over time.

4. The method for identifying electrical loads according to claim 1 or 2, characterized in that, The Hidden Markov Model specifically includes: in, A i h Indicates production line i In the h The probability of state transition within one hour; P( x t i |x t-1 i )for t -1 hour arrives t Time production line i From running status x t-1 i Transition to running state x t i The probability of n; i,h (s1,s2) For production line i In the corresponding observed active power sequence y i The running status before and after the middle of the time from x t-1 i Transferred to x t-1 i Quantity, N For production line i In the observed power sequence y i The total number of observed operational states included above.

5. The method for identifying electrical loads according to claim 1 or 2, characterized in that, The target model specifically also includes: in, P ( x t i | x t-1 i ) indicates the first i The production line is at the h From within hours x t-1 i Running status x t i The transition probability of the running state, P ( y t | x t i ) is when t Time production line i In running status x t i The total observed power is y t The probability of.

6. The method for identifying electrical loads according to claim 5, characterized in that, The target model specifically also includes: in, P ( y t i | x t i )yes t Time production line i In running status x t i The observed active power is y t i The probability of.

7. An electrical load identification device, characterized in that, The device includes: A construction unit is used to construct initial models corresponding to multiple preset time periods. The initial model represents the correlation between the total power of each production line and the various operating states of each production line and the energy distribution of each production line within each preset time period. The training unit is used to optimize the model parameters of each initial model based on the historical load data of each production line in each preset time period, with the goal of maximizing the probability of each historical operating state of each production line under the historical total power of each production line. The historical load data includes a historical active power sequence composed of multiple historical active power of each production line in each preset time period, as well as the number of historical operating states under each historical active power, the historical distribution probability of each historical operating state under each historical active power, and the historical transition probability of each historical operating state of each production line in all production lines switching operating states. The identification unit is used to obtain observed load data from the observed operating data of each production line within a preset time period before the current time and with the smallest time interval to the current time, and input the observed load data into the target model corresponding to the preset time period before the current time and with the smallest time interval to the current time, to obtain the target operating states of each production line under the current total power and the target energy distribution of each production line under each target operating state; wherein, each target operating state is obtained with the goal of maximizing the probability of each operating state to which each production line belongs under the observed total power of each production line; the target energy distribution of each production line under each target operating state is obtained with the goal of maximizing the probability of active power to which each production line belongs under each target operating state of each production line. The target model includes a Gaussian mixture model and a hidden Markov model; the identification unit is specifically used for: The Gaussian mixture distribution model in the target model is invoked to cluster the current observed operating data of each production line, thereby obtaining the number of multiple observed operating states included in each production line and the distribution probability of each production line being in each observed operating state under each observed active power; the multiple observed active powers of each production line constitute an observed active power sequence. Using the Hidden Markov Model in the target model, the transition probability of each observed operating state switching in all production lines is determined based on the switching frequency of different time-adjacent observed operating states in each production line.

8. An electrical device comprising computer instructions that, when executed on the electrical device, cause the electrical device to perform the power load identification method as described in any one of claims 1-6.

9. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions in the computer-readable storage medium are executed by a processor, they enable the electrical load identification method as described in any one of claims 1-6 to be performed.

Images