A power grid dispatching demand response system and method

Abstract

The present invention relates to the technical field of power grid demand response, and discloses a power grid dispatch demand response system and method. The system comprises a state vector generation module, a state sequence generation module, a demand factor extraction module, a factor association analysis module, a weighted feature generation module and a resource dispatch module. The system can generate a state estimation value of a power grid and perform vector encoding based on pre-collected power grid data to obtain a state vector, and then obtain a state sequence through splicing and dimensionality reduction. At the same time, demand factors are extracted from the demand data, and the relationship between the demand factors is calculated through association analysis. The demand factors are weighted according to the factor association to obtain demand weighted features. Finally, a load management strategy is generated based on these features and the state sequence, so that resources of the power grid are dispatched according to the load management strategy, thereby improving the accuracy of the power grid dispatch demand response.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid demand response, and in particular to a power grid dispatching demand response system and method. Background Art

[0002] As the world pays more attention to environmental protection and emission reduction, the carbon emission targets of the power industry are becoming more and more stringent. Improving the accuracy of grid dispatch demand response can optimize the power generation structure, reduce dependence on high-carbon energy (such as coal), and promote the consumption of renewable energy. At the same time, grid load demand is affected by many factors, such as weather changes, holidays, seasonal fluctuations, social and economic activities, etc. The changes in load demand are often uncertain, which brings great challenges to grid dispatch.

[0003] There are complex correlations between demand factors, which may change over time and with changes in the external environment in actual power grids. It is necessary to fully consider demand factors and cope with the complex correlations between demand factors and changes brought about by the ever-changing external environment. Therefore, the power grid dispatching system must have greater flexibility and more accurate demand response capabilities. Summary of the invention

[0004] The present invention provides a power grid dispatching demand response system and method, the main purpose of which is to solve the problem of low accuracy in power grid dispatching demand response.

[0005] To achieve the above object, the present invention provides a power grid dispatch demand response system, which includes a state vector generation module, a state sequence generation module, a demand factor extraction module, a factor association analysis module, a weighted feature generation module and a resource scheduling module, wherein:

[0006] The state vector generating module is used to generate a power grid state estimation value of the power grid according to the pre-collected power grid data, and perform vector encoding on the power grid state estimation value to obtain a state vector of the power grid;

[0007] The state sequence generation module is used to perform vector concatenation on the state vectors to obtain a state matrix of the power grid, and perform feature dimension reduction on the state matrix to obtain a state sequence of the power grid;

[0008] The demand factor extraction module is used to perform parallel calculation on the pre-collected demand data, and extract the demand factor of the power grid according to the calculation results of the parallel calculation;

[0009] The factor correlation analysis module is used to perform correlation analysis on the demand factors using a preset correlation algorithm to obtain the factor correlation of the demand factors. , wherein the preset association algorithm is:

[0010] ;

[0011] in, is the change value under the baseline condition, It is for A weighted index of changes in demand factors, It is for The normalized index of changes in demand factors, It is for A weighted index of changes in demand factors, It is for The normalized index of changes in demand factors, It is with The change value related to each demand factor, It is with The change value related to each demand factor, It is the indicator of demand factors. It is the indicator of demand factors. is the monitoring time point of demand factors, is the number of samples at the monitoring time point;

[0012] The weighted feature generation module is used to perform weighted processing on the demand factors according to the update weights determined by the factor associations to obtain the demand weighted features of the power grid;

[0013] The resource scheduling module is used to generate a load management strategy for the power grid according to the demand weighted characteristics and the state sequence, and to perform resource scheduling for the power grid according to the load management strategy.

[0014] Optionally, when the state vector generation module generates a power grid state estimation value of the power grid according to the pre-collected power grid data, the state vector generation module includes:

[0015] Collecting grid data of the grid according to the pre-acquired grid nodes, wherein the grid data includes voltage, current, electric power, and electric frequency;

[0016] Generate a grid state estimation value of the grid according to a preset state estimation algorithm and the grid data, wherein the preset state estimation algorithm is:

[0017] ;

[0018] in, It is the current moment Grid state estimation, It was the last moment Grid state estimation, is the gain matrix, It is the current moment The measured value of is the observation matrix.

[0019] Optionally, the gain matrix for:

[0020] ;

[0021] in, It was the last moment The state error covariance matrix, is the transpose of the observation matrix, is the observation matrix, is the covariance matrix of the measurement noise.

[0022] Optionally, when the state vector generating module performs vector encoding on the power grid state estimation value to obtain the state vector of the power grid, the state vector generating module includes:

[0023] Determine a current state estimation value, a voltage state estimation value, an electric power state estimation value, and an electric frequency state estimation value of the electric grid according to the electric grid state estimation value;

[0024] The current state estimation value, the voltage state estimation value, the electric power state estimation value and the electric frequency state estimation value are vector-encoded one by one to obtain the state vector of the power grid, wherein the state vector for:

[0025] ;

[0026] in, is the state vector, is the voltage state estimate, is the current state estimate, is the estimated value of the electric power state, is the power frequency state estimate, It is a time stamp.

[0027] Optionally, when the state sequence generation module performs vector concatenation on the state vector to obtain the state matrix of the power grid, the state sequence generation module includes:

[0028] determining a timestamp of the state vector;

[0029] Perform vector concatenation on the state vector according to the timestamp to obtain a state matrix , where the state matrix is:

[0030] ;

[0031] in, Represents the index of different time points, is a vector matrix concatenated according to the state vector, is the total number of time points.

[0032] Optionally, when the state sequence generation module performs feature dimension reduction on the state matrix to obtain the state sequence of the power grid, the module includes:

[0033] Based on the principal component analysis algorithm, the state matrix is ​​subjected to feature dimension reduction to obtain the state sequence of the power grid. , where the state sequence for:

[0034] ;

[0035] in, It is at the moment The state sequence of the power grid, is the total number of data in the state sequence, It is a time stamp.

[0036] Optionally, when the demand factor extraction module performs parallel calculation on the pre-collected demand data and extracts the demand factor of the power grid according to the calculation result of the parallel calculation, it includes:

[0037] Performing quantum encoding on the pre-collected demand data to obtain quantum bit data of the demand data;

[0038] Performing a ground state conversion on the quantum bit data, and performing parallel calculation on the data after the ground state conversion;

[0039] The calculation results of the parallel calculation are quantum measured to obtain the classical results of the required data;

[0040] The key features of the demand data are extracted according to the classical results, and the key features are determined to be demand factors of the power grid, wherein the demand factors are: economic factors, social activity factors and meteorological factors.

[0041] Optionally, when the weighted feature generation module performs weighted processing on the demand factors according to the update weights determined according to the factor relevance to obtain the demand weighted features of the power grid, the weighted feature generation module includes:

[0042] generating an influencing factor of the demand factor according to the factor correlation, and updating a preset demand factor weight according to the influencing factor to obtain an updated weight of the demand factor;

[0043] The demand factors are weighted using the updated weights to obtain the demand weighted characteristics of the power grid.

[0044] Optionally, when the resource scheduling module generates a load management strategy for the power grid according to the demand weighted characteristics and the state sequence, the resource scheduling module includes:

[0045] Establishing a quantitative relationship between the demand weighted feature and the state sequence;

[0046] obtaining real-time demand data of the power grid;

[0047] Determining a target state value of the power grid according to the real-time demand data and the quantitative relationship;

[0048] A load management strategy for the power grid is generated according to the target state value and pre-acquired distributed resource data.

[0049] In order to solve the above problems, the present invention further provides a power grid dispatching demand response method, the method comprising:

[0050] Generate a power grid state estimation value of the power grid according to the pre-collected power grid data, and perform vector encoding on the power grid state estimation value to obtain a state vector of the power grid;

[0051] Performing vector concatenation on the state vectors to obtain a state matrix of the power grid, and performing feature dimension reduction on the state matrix to obtain a state sequence of the power grid;

[0052] Performing parallel calculation on the pre-collected demand data, and extracting demand factors of the power grid according to the calculation results of the parallel calculation;

[0053] Use a preset correlation algorithm to perform correlation analysis on the demand factors to obtain the factor correlation of the demand factors , wherein the preset association algorithm is:

[0054] ;

[0055] in, is the change value under the baseline condition, It is for A weighted index of changes in demand factors, It is for The normalized index of changes in demand factors, It is for A weighted index of changes in demand factors, It is for The normalized index of changes in demand factors, It is with The change value related to each demand factor, It is with The change value related to each demand factor, It is the indicator of demand factors. It is the indicator of demand factors. is the monitoring time point of demand factors, is the number of samples at the monitoring time point;

[0056] The demand factors are weighted according to the updated weights determined by the correlation of the factors to obtain the demand weighted characteristics of the power grid;

[0057] A load management strategy for the power grid is generated according to the demand weighted characteristics and the state sequence, and resources of the power grid are scheduled according to the load management strategy.

[0058] The present invention generates an estimated value of the state of the power grid through real-time collection of power grid data, and vectorizes it to provide an accurate representation of the power grid state; the state vectors are spliced ​​to form a state matrix of the power grid, and feature dimension reduction is performed to obtain the state sequence of the power grid, thereby reducing unnecessary data redundancy; in large-scale power grids, when a large amount of demand data needs to be extracted quickly, traditional serial calculations often become a bottleneck, and parallel calculations can shorten the calculation time by dividing the data into blocks and processing them simultaneously, so that demand factors can be extracted in time, providing real-time support for scheduling decisions; the preset association algorithm quantifies the impact of changes in different demand factors on the power grid load by comparing the changes in each demand factor with the baseline state, and reveals the correlation analysis of different demand factors through correlation analysis. At the same time, a differentiated approach (i.e., calculating the difference in changes at each monitoring time point) is used to reveal the relationships between different demand factors, and the grid dispatching strategy is optimized according to the historical change trend to improve the accuracy of dispatching; on the basis of factor association analysis, a demand weighted feature is generated, and the grid dispatching is made more accurate by adjusting the weights of demand factors. By comprehensively considering the real-time state and demand characteristics of the grid, grid resources can be reasonably allocated under various load fluctuation situations, power distribution can be optimized, and energy waste can be reduced. The combination of weighted features and state sequences makes the load management strategy more real-time and adaptable. Therefore, the grid dispatching demand response system and method proposed in the present invention can improve the accuracy of grid dispatching demand response. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A system architecture diagram of a power grid dispatching demand response system provided by an embodiment of the present invention;

[0060] Figure 2 A schematic flow chart of a method for responding to power grid dispatching demand provided in one embodiment of the present invention.

[0061] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings, and "multiple" generally includes at least two.

[0064] As used herein, the words "if" and "if" may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0065] In addition, the sequence of steps in the following method embodiments is only an example and not a strict limitation.

[0066] In fact, the service-side device deployed by the power grid dispatching demand response system may be composed of one or more devices. The above-mentioned power grid dispatching demand response system can be implemented as: business instance, virtual machine, hardware device. For example, the power grid dispatching demand response system can be implemented as a business instance deployed on one or more devices in the cloud node. In simple terms, the power grid dispatching demand response system can be understood as a software deployed on the cloud node, which is used to provide a power grid dispatching demand response system for each user end. Alternatively, the power grid dispatching demand response system can also be implemented as a virtual machine deployed on one or more devices in the cloud node. The virtual machine is installed with application software for managing each user end. Alternatively, the power grid dispatching demand response system can also be implemented as a server composed of many hardware devices of the same or different types, and one or more hardware devices are set to provide a power grid dispatching demand response system for each user end.

[0067] In terms of implementation, the power grid dispatching demand response system and the user end are adapted to each other. That is, the power grid dispatching demand response system is an application installed on the cloud service platform, and the user end is a client that establishes a communication connection with the application; or the power grid dispatching demand response system is implemented as a website, and the user end is implemented as a web page; or the power grid dispatching demand response system is implemented as a cloud service platform, and the user end is implemented as a small program in an instant messaging application.

[0068] like Figure 1 , which is a system architecture diagram of a power grid dispatching demand response system provided in one embodiment of the present invention.

[0069] The power grid dispatching demand response system 100 of the present invention can be set in a cloud server. In terms of implementation, it can be used as one or more service devices, or it can be installed as an application on the cloud (such as a server of a mobile service operator, a server cluster, etc.), or it can be developed as a website. According to the functions implemented, the power grid dispatching demand response system 100 can include a state vector generation module 101, a state sequence generation module 102, a demand factor extraction module 103, a factor association analysis module 104, a weighted feature generation module 105 and a resource scheduling module 106. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0070] In the embodiment of the present invention, in the power grid dispatching demand response system, each of the above modules can be implemented independently and called with other modules. The call here can be understood as a module that can connect to multiple modules of another type and provide corresponding services to the multiple modules connected to it. For example, the sharing evaluation module can call the same information acquisition module to obtain the information collected by the information acquisition module. Based on the above characteristics, in the power grid dispatching demand response system provided by the embodiment of the present invention, the scope of application of the power grid dispatching demand response system architecture can be adjusted by adding modules and directly calling them without modifying the program code, so as to achieve cluster-based horizontal expansion, so as to achieve the purpose of quickly and flexibly expanding the power grid dispatching demand response system. In actual applications, the above modules can be set in the same device or different devices, or they can be set in virtual devices, such as service instances in cloud servers.

[0071] The following describes the various components and specific workflows of the power grid dispatching demand response system in conjunction with specific embodiments:

[0072] The state vector generating module 101 is used to generate a power grid state estimation value of the power grid according to pre-collected power grid data, and perform vector encoding on the power grid state estimation value to obtain a state vector of the power grid.

[0073] In an embodiment of the present invention, when the state vector generation module generates a power grid state estimation value of the power grid according to the pre-collected power grid data, it includes:

[0074] Collecting grid data of the grid according to the pre-acquired grid nodes, wherein the grid data includes voltage, current, electric power, and electric frequency;

[0075] Generate a grid state estimation value of the grid according to a preset state estimation algorithm and the grid data, wherein the preset state estimation algorithm is:

[0076] ;

[0077] in, It is the current moment Grid state estimation, It was the last moment Grid state estimation, is the gain matrix, It is the current moment The measured value of is the observation matrix.

[0078] In detail, it is assumed that the following data has been collected from the power grid (assumed to be measurement data at a certain time k): voltage (V) is 380V, current (I) is 50A, electric power (P) is 10kW, and power frequency (f) is 50Hz.

[0079] Furthermore, according to the given state estimation algorithm, assuming that the last moment is known Estimated value of and measured values , through the gain matrix Estimate the current state and finally calculate the current moment The estimated value of the power grid status.

[0080] In detail, the gain matrix for:

[0081] ;

[0082] in, It was the last moment The state error covariance matrix, is the transpose of the observation matrix, is the observation matrix, is the covariance matrix of the measurement noise.

[0083] In detail, assume that the state error covariance matrix at the previous moment is for:

[0084] ;

[0085] Observation Matrix for:

[0086] ;

[0087] Measurement noise covariance matrix for:

[0088] ;

[0089] Assuming that the voltage and current at the previous moment are 380V and 50A respectively, the state estimation value at the previous moment is for:

[0090] ;

[0091] Assuming that the voltage and current at the current moment are 385V and 52A respectively, the measured values ​​at the current moment are for:

[0092] ;

[0093] Furthermore, the gain matrix is ​​calculated , substituting these data into the formula, first calculate for:

[0094] ;

[0095] because is the identity matrix, so ,so:

[0096] ;

[0097] Next, calculate :

[0098] ;

[0099] Then calculate the gain matrix :

[0100] ;

[0101] In detail, the gain matrix is ​​a diagonal matrix, each diagonal element represents the gain of a state variable, in this example, it corresponds to the gain of two state variables of the power grid (assuming voltage and current).

[0102] The gain value of 0.7143 indicates the degree of response of the system to the observed data. The closer it is to 1, the more the system trusts the current observed data. Conversely, the system trusts the predicted value more. In this example, the gains of voltage and current are the same, indicating that the system gives the same weight to the observed data of voltage and current.

[0103] Furthermore, according to the formula, we can finally calculate Matrix, substitute it into the preset state estimation algorithm to obtain the state estimation value of the power grid.

[0104] First, calculate :

[0105] ;

[0106] Next, the state estimate update is calculated:

[0107] ;

[0108] Therefore, the estimated state of the power grid is:

[0109] ;

[0110] Among them, 383.57 is the estimated voltage value in volts (V), and 51.43 represents the estimated current value in amperes (A). These values ​​are the best estimate of the grid state after combining the system prediction (the state at the last moment) and the current measurement (such as the output of the voltage and current sensors).

[0111] In detail, the gain matrix describes the weight distribution when the observed value is combined with the predicted value. The larger the value, the more the system trusts the observed data, and the smaller the value, the more the system trusts the predicted data; the state estimate is the system's best estimate of the grid state (such as voltage and current) at the current moment, which combines the system's prediction and current measurement.

[0112] Similarly, when the observed values ​​include electric power and electric frequency, the same calculation method is used.

[0113] In the embodiment of the present invention, when the state vector generation module performs vector encoding on the power grid state estimation value to obtain the state vector of the power grid, it includes:

[0114] Determine a current state estimation value, a voltage state estimation value, an electric power state estimation value, and an electric frequency state estimation value of the electric grid according to the electric grid state estimation value;

[0115] The current state estimation value, the voltage state estimation value, the electric power state estimation value and the electric frequency state estimation value are vector-encoded one by one to obtain the state vector of the power grid, wherein the state vector for:

[0116] ;

[0117] in, is the state vector, is the voltage state estimate, is the current state estimate, is the estimated value of the electric power state, is the power frequency state estimate, It is a time stamp.

[0118] In the embodiment of the present invention, various estimated values ​​of the state of the power grid (current, voltage, electric power and electric frequency) may be integrated into a state vector.

[0119] For example, assuming that the following power grid states have been estimated by the state estimation algorithm: voltage is 380 V; current is 50 A; power is 10 kW; power frequency is 50 Hz, then the state vector of the power grid is:

[0120] ;

[0121] The state sequence generating module 102 is used to perform vector concatenation on the state vectors to obtain a state matrix of the power grid, and perform feature dimension reduction on the state matrix to obtain a state sequence of the power grid.

[0122] In the embodiment of the present invention, when the state sequence generation module performs vector concatenation on the state vector to obtain the state matrix of the power grid, the state sequence generation module includes:

[0123] determining a timestamp of the state vector;

[0124] Perform vector concatenation on the state vector according to the timestamp to obtain a state matrix , where the state matrix is:

[0125] ;

[0126] in, Represents the index of different time points, is a vector matrix concatenated according to the state vector, is the total number of time points.

[0127] In the embodiment of the present invention, it is assumed that at multiple time points ( , , , , , ) on the grid, where the different state vectors of the grid are , , ) correspond to three state vectors respectively, and the state vectors are vector-joined to obtain the state matrix of the power grid. The state matrix X formed by joining these state vectors is:

[0128] ;

[0129] Among them, at three time points ( , , ) correspond to three voltage state estimation values ​​380, 385, 379, three current state estimation values ​​50, 52, 51, three power state estimation values ​​10, 12, 11, and three power frequency state estimation values ​​50, 49.9, 50.1

[0130] In the embodiment of the present invention, when the state sequence generation module performs feature dimension reduction on the state matrix to obtain the state sequence of the power grid, it includes:

[0131] Based on the principal component analysis algorithm, the state matrix is ​​subjected to feature dimension reduction to obtain the state sequence of the power grid. , where the state sequence for:

[0132] ;

[0133] in, It is at the moment The state sequence of the power grid, is the total number of data in the state sequence, It is a time stamp.

[0134] The demand factor extraction module 103 is used to perform parallel calculation on the pre-collected demand data, and extract the demand factor of the power grid according to the calculation result of the parallel calculation.

[0135] In the embodiment of the present invention, when the demand factor extraction module performs parallel calculation on the pre-collected demand data and extracts the demand factor of the power grid according to the calculation result of the parallel calculation, it includes:

[0136] Performing quantum encoding on the pre-collected demand data to obtain quantum bit data of the demand data;

[0137] Performing a ground state conversion on the quantum bit data, and performing parallel calculation on the data after the ground state conversion;

[0138] The calculation results of the parallel calculation are quantum measured to obtain the classical results of the required data;

[0139] The key features of the demand data are extracted according to the classical results, and the key features are determined to be demand factors of the power grid, wherein the demand factors are: economic factors, social activity factors and meteorological factors.

[0140] In detail, through parallel computing, it is possible to quickly analyze the changes in multiple demand factors and respond according to different environmental conditions. This enables the demand response system to more quickly identify load change trends and take timely measures.

[0141] In detail, temperature changes may affect the use load of air conditioners and electric heaters at the same time. Changes in economic factors can timely capture the upward or downward trend of power load through forecasts of productivity, market activities, etc. For example, on the eve of holidays or important festivals, the power demand of retail and shopping malls may surge. Through the collection and analysis of real-time economic data, the power grid dispatching system can accurately respond to the fluctuations in power demand caused by economic activities, optimize the load dispatching of the power grid, and avoid overload.

[0142] Specifically, social activity factors refer to the impact of social behaviors and events on electricity demand, such as holidays, major social events, large conferences or sports events, etc. Social activities not only affect people's living habits, but may also significantly change the demand pattern of electricity.

[0143] In detail, changes in economic factors can timely capture the upward or downward trend of electricity load through forecasts of productivity, market activities, etc. For example, on the eve of holidays or important festivals, the electricity demand of retail and shopping malls may surge.

[0144] In detail, it is assumed that the following demand data have been collected (for example, meteorological factors, economic factors, social activity factors, etc.):

[0145] ;

[0146] In detail, assuming that quantum bits are used to represent the required data, each data point can be converted into a corresponding binary sequence and encoded by the quantum bits.

[0147] For example, the economic factor 300 can be encoded as a qubit state , social activity factor 20 is coded as , the temperature 22°C is coded as ,etc.

[0148] In detail, for each demand data (such as temperature, economic and social activity factors), quantum computing will be performed, and these quantum bits will be processed in parallel to perform superposition and interference computing of states. This means that data of multiple demand factors will be processed simultaneously, rather than one by one. For example, for economic factor data 300, quantum computing may generate superposition states to represent multiple possible economic states. Similarly, the quantum states of social activity factors and meteorological factors will also be processed in parallel computing.

[0149] In detail, after the calculation is completed, quantum measurement is performed to obtain the classical result. Here, by measuring the state of the quantum bit, the classical demand data result is obtained, for example: the classical result obtained after measurement is economic factor 310, social activity factor 22, and meteorological factor 24℃. Key features such as economic factors, social activity factors, and meteorological factors are extracted from the classical results. These extracted features will constitute the demand factors of the power grid. For example: the economic factor is 310, the social activity factor is 22, and the meteorological factor (temperature) is 24℃.

[0150] In detail, through quantum computing parallel processing, multiple demand data (for example, temperature, economic activities, social factors) can be processed simultaneously, and through quantum measurement, reliable classical data can be extracted from the results of multiple parallel calculations. These data serve as demand factors for the power grid. The role of parallel computing here is to improve computing efficiency and processing capabilities, while ensuring that the data of each demand factor can be integrated together, thereby achieving more accurate demand factor extraction.

[0151] The factor correlation analysis module 104 is used to perform correlation analysis on the demand factors using a preset correlation algorithm to obtain the factor correlation of the demand factors. .

[0152] In the embodiment of the present invention, the preset association algorithm is:

[0153] ;

[0154] in, is the change value under the baseline condition, It is for A weighted index of changes in demand factors, It is for The normalized index of changes in demand factors, It is for A weighted index of changes in demand factors, It is for The normalized index of changes in demand factors, It is with The change value related to each demand factor, It is with The change value related to each demand factor, It is the indicator of demand factors. It is the indicator of demand factors. is the monitoring time point of demand factors, is the number of samples at the monitoring time point.

[0155] In an embodiment of the present invention, The value is the change of the reference benchmark at each monitoring time point The value of and They are respectively The first demand factor and the The change value related to the demand factor at the time point Changes in demand factors, and Respectively for and A weighted index of changes in demand factors, usually used to indicate the weight of these factors on the final calculation result; and Respectively for and A normalized index of changes in demand factors, indicating how to normalize the weights of these changes in calculations; is the number of samples at a time point, that is, how many time points of data need to be considered.

[0156] In detail, suppose we have the following demand factor data, representing changes at different points in time:

[0157] ;

[0158] Furthermore, assume that the following parameters are used:

[0159] ;

[0160] In detail, calculate and

[0161] Regarding social activities and baseline status :

[0162] 08:00, ;

[0163] 12:00, ;

[0164] 16:00, ;

[0165] 20:00, ;

[0166] Furthermore, the weighted sum of economic factors:

[0167] ;

[0168] Weighted sum of social activity factors:

[0169] ;

[0170] Normalized for economic factors and:

[0171] ;

[0172] Normalization of social activity factors and;

[0173] ;

[0174] Further, substitute the formula to calculate the correlation for:

[0175] ;

[0176] Further, we calculate ,This result shows that the correlation between economic factors and social activity factors is 266.64, ,which can be used to reflect the relative changes of two demand factors at different ,time points.

[0177] In detail, the correlation The higher the value, the more similar the two demand factors are in the process of change. The weighted index in the formula makes different factors have different weights in the calculation, and the normalized index helps to balance the differences in magnitude between different factors and avoid some factors dominating the correlation calculation due to their large magnitude.

[0178] In detail, correlation analysis can reveal the mutual influence between demand factors, so as to more accurately predict demand changes. For example, temperature changes may affect the usage load of air conditioners and electric heaters at the same time. Through accurate correlation analysis, load management strategies can be adjusted to more effectively regulate grid load.

[0179] The weighted feature generation module 105 is used to perform weighted processing on the demand factors according to the updated weights determined by the factor associations to obtain the demand weighted features of the power grid.

[0180] In the embodiment of the present invention, when the weighted feature generation module performs weighted processing on the demand factors according to the update weights determined according to the factor relevance to obtain the demand weighted features of the power grid, the weighted feature generation module includes:

[0181] generating an influencing factor of the demand factor according to the factor correlation, and updating a preset demand factor weight according to the influencing factor to obtain an updated weight of the demand factor;

[0182] The demand factors are weighted using the updated weights to obtain the demand weighted characteristics of the power grid.

[0183] In detail, the analysis of the correlation between different demand factors is completed by calculating the relationship between different factors (the correlation formula mentioned above). The more strongly correlated the factor is, the greater its impact on the power grid may be, so it needs to be given a higher weight.

[0184] In detail, the generation of weighted features ensures that changes in demand factors can be appropriately adjusted and responded to according to their impact on the grid load. By accurately weighting demand factors, the load management strategy can make more precise adjustments in different time periods according to different load demands, ensuring the balance and efficiency of the grid.

[0185] Further, updating the weight is to adjust the relative importance of each demand factor according to the calculated correlation. Generally, the higher the correlation, the closer the relationship between the two demand factors, so their weights will increase accordingly.

[0186] Assume that the weights of the original demand factors are and , then the updated weights can be calculated as follows:

[0187] ;

[0188] In this way, the updated weights reflect the similarity of the changes of the two demand factors at different time points. The larger it is, the more similar the changes between the two demand factors are, and their weights will increase accordingly.

[0189] In detail, the impact factor is a comprehensive indicator used to measure the impact of demand factors on the overall performance of the power grid system. To determine the influencing factor of each demand factor. Specifically, the influencing factor can be expressed as:

[0190] ;

[0191] Among them, among them, and They are the influencing factors related to economic factors and social activity factors, Indicates the use of association and weight The calculated impact factor will be used to update the weight of the demand factor.

[0192] Furthermore, according to the calculated impact factor and , we adjust the weight of demand factors. Update using the following formula:

[0193] ;

[0194] In this way, the weights of demand factors can be updated according to their relevance, influencing factors, and roles in the grid system.

[0195] In detail, the updated demand factor weights will be used to calculate the weighted characteristics of the power grid. The demand weighted characteristics of the power grid can be obtained by weighted calculation of the weights of each demand factor and the corresponding characteristic data.

[0196] For example, if the demand characteristics of the power grid include multiple factors , then the demand weighted characteristics of the power grid can be expressed as:

[0197] ;

[0198] in, is the updated weight, is a characteristic of each demand factor.

[0199] In general, by calculating the correlation, we can obtain the correlation between demand factors and adjust the weights of demand factors based on this. Then, we generate the impact factors and further adjust the weights to obtain the demand weighted characteristics of the power grid. This process effectively evaluates and reflects the importance and influence of different demand factors in the power grid by means of weighting and standardization.

[0200] The resource scheduling module 106 is used to generate a load management strategy for the power grid according to the demand weighted characteristics and the state sequence, and perform resource scheduling for the power grid according to the load management strategy.

[0201] In an embodiment of the present invention, when the resource scheduling module generates a load management strategy for the power grid according to the demand weighted characteristics and the state sequence, the resource scheduling module includes:

[0202] Establishing a quantitative relationship between the demand weighted feature and the state sequence;

[0203] obtaining real-time demand data of the power grid;

[0204] Determining a target state value of the power grid according to the real-time demand data and the quantitative relationship;

[0205] A load management strategy for the power grid is generated according to the target state value and pre-acquired distributed resource data.

[0206] In detail, a quantitative relationship is established between the weighted characteristics of grid demand (such as meteorology, social activities, economic factors, etc.) and the state sequence (such as current state, voltage state, electric power state, and power frequency state, etc.). This can be done through data-driven methods, such as regression analysis, machine learning or deep learning models, in order to provide accurate predictions for the load management strategy of the grid.

[0207] Specifically, the weighted environmental characteristics can accurately reflect the actual impact of various factors on the power grid, and the load management strategy will consider these weighted characteristics to formulate a dispatch plan. For example, if a certain environmental factor (such as temperature) has a greater impact on the power grid load, the system can dynamically dispatch according to this weighted characteristic.

[0208] In detail, assume that the demand weighting characteristics are , the state sequence is , we can establish the following quantitative relationship:

[0209] ;

[0210] in, It is a mapping relationship from demand weighted features to state sequences. This mapping relationship can be constructed through a variety of machine learning algorithms (such as regression analysis, neural networks, support vector machines, etc.).

[0211] In detail, real-time demand data refers to external demand information that can reflect the operation status of the power grid at the current moment, such as weather data, economic data, social activity data, etc. These data will be used as input for load management strategies to update the power grid status in real time.

[0212] In detail, after obtaining the real-time demand data, the target state value of the power grid is calculated by combining it with the quantitative relationship established in the first step. This is the target state that the power grid should achieve at the current point in time, which may include target values ​​of key indicators such as power grid load, frequency, and voltage.

[0213] In detail, the core goal of the load management strategy is to adjust the load and resource usage of the power grid according to the target state value to ensure that the power grid operates stably and meets the demand.

[0214] In detail, it is necessary to obtain data on distributed energy resources, including but not limited to: distributed power generation data (such as power generation of solar photovoltaic, wind power, etc.); energy storage equipment data (such as the status of battery energy storage, chargeable and discharge capacity, etc.); demand-side management data (such as demand response capabilities of commercial and household users).

[0215] Furthermore, load management strategies can be generated through optimization algorithms (such as linear programming, dynamic programming, or genetic algorithms), with the goal of minimizing the operating cost of the power grid or maximizing system stability.

[0216] In detail, through accurate load management strategies, the power grid can adjust the distribution of electricity in a timely manner to avoid overload or power shortage, and ensure the smooth operation of the power grid. In actual operation, the load management strategy can be dynamically adjusted according to the real-time status of the power grid and changes in demand, thereby improving the accuracy and reliability of power grid dispatching.

[0217] In detail, for example, according to the real-time load demand of the power grid, the load pressure of the power grid can be balanced by incentivizing users to increase electricity consumption during low-demand periods and reduce electricity consumption during high-demand periods; or by predicting users' electricity demand, combining power production and distribution capabilities, and reasonably dispatching power generation and distribution resources to ensure stable power supply.

[0218] In detail, the core of demand response is to respond to the load changes of the power grid and the influence of external environmental factors through dynamic adjustment of user power demand. Through the dispatch strategy, the system can flexibly implement load reduction or load transfer for different regions and different user groups, so as to achieve balance and optimization of the power grid.

[0219] In detail, when the load management strategy is activated, resource scheduling will automatically take these changes into account and adjust the grid's operating strategy accordingly, which may include charging and discharging control of energy storage devices, flexible scheduling of different types of generators, and deployment of other backup resources to ensure that the grid can have sufficient power supply at all times.

[0220] In detail, the load management strategy reflects the arrangement of resource scheduling, which transmits electricity from the power generation side to each load point through the distribution system. Resource scheduling dynamically adjusts the working status of each power grid link to ensure that power resources can be provided efficiently and stably at any time.

[0221] In general, a reasonable dispatching strategy is generated according to the demand characteristics and the grid status, so as to achieve effective dispatching of grid resources, avoid overload and improve the reliability and efficiency of the grid.

[0222] Reference Figure 2 FIG. 1 is a flow chart of a method for responding to power grid dispatching demand provided by an embodiment of the present invention. In this embodiment, the method for responding to power grid dispatching demand includes:

[0223] Generate a power grid state estimation value of the power grid according to the pre-collected power grid data, and perform vector encoding on the power grid state estimation value to obtain a state vector of the power grid;

[0224] Performing vector concatenation on the state vectors to obtain a state matrix of the power grid, and performing feature dimension reduction on the state matrix to obtain a state sequence of the power grid;

[0225] Performing parallel calculation on the pre-collected demand data, and extracting demand factors of the power grid according to the calculation results of the parallel calculation;

[0226] Use a preset correlation algorithm to perform correlation analysis on the demand factors to obtain the factor correlation of the demand factors , wherein the preset association algorithm is:

[0227] ;

[0228] in, is the change value under the baseline condition, It is for A weighted index of changes in demand factors, It is for The normalized index of changes in demand factors, It is for A weighted index of changes in demand factors, It is for The normalized index of changes in demand factors, It is with The change value related to each demand factor, It is with The change value related to each demand factor, It is the indicator of demand factors. It is the indicator of demand factors. is the monitoring time point of demand factors, is the number of samples at the monitoring time point;

[0229] The demand factors are weighted according to the updated weights determined by the correlation of the factors to obtain the demand weighted characteristics of the power grid;

[0230] A load management strategy for the power grid is generated according to the demand weighted characteristics and the state sequence, and resources of the power grid are scheduled according to the load management strategy.

[0231] The present invention generates an estimated value of the state of the power grid through real-time collection of power grid data, and vectorizes it to provide an accurate representation of the power grid state; the state vectors are spliced ​​to form a state matrix of the power grid, and feature dimension reduction is performed to obtain the state sequence of the power grid, thereby reducing unnecessary data redundancy; in large-scale power grids, when a large amount of demand data needs to be extracted quickly, traditional serial calculations often become a bottleneck, and parallel calculations can shorten the calculation time by dividing the data into blocks and processing them simultaneously, so that demand factors can be extracted in time, providing real-time support for scheduling decisions; the preset association algorithm quantifies the impact of changes in different demand factors on the power grid load by comparing the changes in each demand factor with the baseline state, and reveals the correlation analysis of different demand factors through correlation analysis. The relationship between factors, at the same time, a differentiated approach (i.e., calculating the difference in changes at each monitoring time point) is used to reveal the relationship between different demand factors, and the power grid dispatching strategy is optimized according to the historical change trend to improve the accuracy of dispatching; on the basis of factor association analysis, a demand weighted feature is generated, and the power grid dispatching is made more accurate by adjusting the weights of the demand factors. By comprehensively considering the real-time state and demand characteristics of the power grid, the power grid resources can be reasonably allocated under various load fluctuation situations, the power distribution can be optimized and energy waste can be reduced. The combination of weighted features and state sequences makes the load management strategy more real-time and adaptable. Therefore, the power grid dispatching demand response method proposed in the present invention can improve the accuracy of power grid dispatching demand response.

[0232] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0233] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0234] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive needs, acquire knowledge and use knowledge to obtain the best results.

[0235] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in the system claim can also be implemented by one unit or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0236] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A power grid dispatching demand response system, characterized in that: The system includes a state vector generation module, a state sequence generation module, a demand factor extraction module, a factor association analysis module, a weighted feature generation module and a resource scheduling module, wherein: The state vector generating module is used to generate a power grid state estimation value of the power grid according to the pre-collected power grid data, and perform vector encoding on the power grid state estimation value to obtain a state vector of the power grid; The state sequence generation module is used to perform vector concatenation on the state vectors to obtain a state matrix of the power grid, and perform feature dimension reduction on the state matrix to obtain a state sequence of the power grid; The demand factor extraction module is used to perform parallel calculation on the pre-collected demand data, and extract the demand factor of the power grid according to the calculation results of the parallel calculation; The factor correlation analysis module is used to perform correlation analysis on the demand factors using a preset correlation algorithm to obtain the factor correlation of the demand factors. , wherein the preset association algorithm is: ； in, is the change value under the baseline condition, It is for A weighted index of changes in demand factors, It is for The normalized index of changes in demand factors, It is for A weighted index of changes in demand factors, It is for The normalized index of changes in demand factors, It is with The change value related to each demand factor, It is with The change value related to each demand factor, It is the indicator of demand factors. It is the indicator of demand factors. is the monitoring time point of demand factors, is the number of samples at the monitoring time point; The weighted feature generation module is used to perform weighted processing on the demand factors according to the update weights determined by the factor associations to obtain the demand weighted features of the power grid; The resource scheduling module is used to generate a load management strategy for the power grid according to the demand weighted characteristics and the state sequence, and to perform resource scheduling for the power grid according to the load management strategy.

2. The power grid dispatching demand response system according to claim 1, characterized in that: When the state vector generation module generates a power grid state estimation value of the power grid according to the pre-collected power grid data, the state vector generation module includes: Collecting grid data of the grid according to the pre-acquired grid nodes, wherein the grid data includes voltage, current, electric power, and electric frequency; Generate a grid state estimation value of the grid according to a preset state estimation algorithm and the grid data, wherein the preset state estimation algorithm is: ； in, It is the current moment Grid state estimation, It was the last moment Grid state estimation, is the gain matrix, It is the current moment The measured value of is the observation matrix.

3. The power grid dispatching demand response system according to claim 2, characterized in that: The gain matrix for: ； in, It was the last moment The state error covariance matrix, is the transpose of the observation matrix, is the observation matrix, is the covariance matrix of the measurement noise.

4. The power grid dispatching demand response system according to claim 1, characterized in that: When the state vector generation module performs vector encoding on the power grid state estimation value to obtain the state vector of the power grid, the state vector generation module includes: Determine a current state estimation value, a voltage state estimation value, an electric power state estimation value, and an electric frequency state estimation value of the electric grid according to the electric grid state estimation value; The current state estimation value, the voltage state estimation value, the electric power state estimation value and the electric frequency state estimation value are vector-encoded one by one to obtain the state vector of the power grid, wherein the state vector for: ； in, is the state vector, is the voltage state estimate, is the current state estimate, is the estimated value of the electric power state, is the power frequency state estimate, Is a time stamp.

5. The power grid dispatching demand response system according to claim 1, characterized in that: When the state sequence generation module performs vector concatenation on the state vector to obtain the state matrix of the power grid, the state sequence generation module includes: determining a timestamp of the state vector; Perform vector concatenation on the state vector according to the timestamp to obtain a state matrix , where the state matrix is: ； in, Represents the index of different time points, is a vector matrix concatenated according to the state vector, is the total number of time points.

6. The power grid dispatching demand response system according to claim 1, characterized in that: When the state sequence generation module performs feature dimension reduction on the state matrix to obtain the state sequence of the power grid, it includes: Based on the principal component analysis algorithm, the state matrix is ​​subjected to feature dimension reduction to obtain the state sequence of the power grid. , where the state sequence for: ； in, It is at the moment The state sequence of the power grid, is the total number of data in the state sequence, It is a time stamp.

7. The power grid dispatching demand response system according to claim 1, characterized in that: The demand factor extraction module, when performing parallel calculation on the pre-collected demand data and extracting the demand factor of the power grid according to the calculation result of the parallel calculation, includes: Performing quantum encoding on the pre-collected demand data to obtain quantum bit data of the demand data; Performing a ground state conversion on the quantum bit data, and performing parallel calculation on the data after the ground state conversion; The calculation results of the parallel calculation are quantum measured to obtain the classical results of the required data; The key features of the demand data are extracted according to the classical results, and the key features are determined to be demand factors of the power grid, wherein the demand factors are: economic factors, social activity factors and meteorological factors.

8. The power grid dispatching demand response system according to claim 1, characterized in that: When the weighted feature generation module performs weighted processing on the demand factors according to the update weights determined according to the factor association to obtain the demand weighted features of the power grid, the weighted feature generation module includes: generating an influencing factor of the demand factor according to the factor correlation, and updating a preset demand factor weight according to the influencing factor to obtain an updated weight of the demand factor; The demand factors are weighted using the updated weights to obtain the demand weighted characteristics of the power grid.

9. The power grid dispatching demand response system according to any one of claims 1 to 8, characterized in that: When the resource scheduling module generates a load management strategy for the power grid according to the demand weighted characteristics and the state sequence, the resource scheduling module includes: Establishing a quantitative relationship between the demand weighted feature and the state sequence; obtaining real-time demand data of the power grid; Determining a target state value of the power grid according to the real-time demand data and the quantitative relationship; A load management strategy for the power grid is generated according to the target state value and pre-acquired distributed resource data.

10. A method for responding to power grid dispatching demand, characterized in that: The method comprises: S1. Generate a power grid state estimation value of a power grid according to pre-collected power grid data, and perform vector encoding on the power grid state estimation value to obtain a state vector of the power grid; S2. performing vector concatenation on the state vectors to obtain a state matrix of the power grid, and performing feature dimension reduction on the state matrix to obtain a state sequence of the power grid; S3, performing parallel calculation on the pre-collected demand data, and extracting the demand factors of the power grid according to the calculation results of the parallel calculation; S4. Perform correlation analysis on the demand factors using a preset correlation algorithm to obtain the factor correlation of the demand factors. , wherein the preset association algorithm is: ； in, is the change value under the baseline condition, It is for A weighted index of changes in demand factors, It is for The normalized index of changes in demand factors, It is for A weighted index of changes in demand factors, It is for The normalized index of changes in demand factors, It is with The change value related to each demand factor, It is with The change value related to each demand factor, It is the indicator of demand factors. It is the indicator of demand factors. is the monitoring time point of demand factors, is the number of samples at the monitoring time point; S5. performing weighted processing on the demand factors according to the updated weights determined by the correlation of the factors to obtain the demand weighted characteristics of the power grid; S6. Generate a load management strategy for the power grid according to the demand weighted characteristics and the state sequence, and perform resource scheduling for the power grid according to the load management strategy.

Images