A power grid intelligent scheduling method and system fusing RPA and AI

Abstract

The application provides a power grid intelligent scheduling method and system fusing RPA and AI, relates to the technical field of data processing, and the method comprises the following steps: constructing a power grid intelligent scheduling model based on artificial intelligence AI, wherein the power grid intelligent scheduling model specifically comprises an energy use behavior prediction module, a power grid scheduling scheme creation module and a scheduling scheme determination module; using a three-layer stacked long short-term memory neural network to predict energy use behavior through the energy use behavior prediction module; using a whale optimization algorithm to create a suboptimal power grid scheduling scheme according to the predicted energy use behavior through the power grid scheduling scheme creation module; determining a target scheduling scheme from the suboptimal power grid scheduling scheme based on a user habit deviation degree through the scheduling scheme determination module; and executing the target scheduling scheme through robot process automation RPA technology.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a power grid intelligent scheduling method and system fusing RPA and AI. BACKGROUND

[0002] Power grid intelligent scheduling can quickly respond to load changes, prevent power grid overload or imbalance, and ensure stable operation of the power system. By optimizing the scheduling scheme, the loss in the process of power generation and transmission can be reduced, and the operating cost can be reduced. Power grid intelligent scheduling is of great significance to the efficient, safe and economic operation of the power system, and promotes the modernization and intelligentization process of energy management.

[0003] Currently, there are many power grid scheduling schemes based on artificial intelligence, such as using genetic algorithm, particle swarm optimization algorithm, ant colony algorithm, etc. to determine the best scheduling scheme, to improve the intelligent level and optimization effect of power grid scheduling.

[0004] However, the current power grid scheduling scheme often only focuses on reducing electricity costs, ignoring the consideration of user electricity habits, which may lead to insufficient power supply for users during high demand periods, affecting daily life and work. For example, if the power supply is insufficient during the summer evening peak period, it will affect the use of air conditioners, lighting, cooking and other habitual electricity demand, resulting in a decrease in user satisfaction.

[0005] At the same time, after the current power grid scheduling scheme is determined, it still needs to be manually executed, which requires a lot of time and manpower. Manual operation inevitably leads to omissions and errors, which may result in inaccurate execution of the scheduling scheme, and even cause power grid failure or power outage. SUMMARY

[0006] In order to solve the technical problems that the current power grid scheduling scheme often only focuses on reducing electricity costs, ignoring the consideration of user electricity habits, which may lead to insufficient power supply for users during high demand periods, affecting daily life and work, and the current power grid scheduling scheme still needs to be manually executed after being determined, which requires a lot of time and manpower, and manual operation inevitably leads to omissions and errors, which may result in inaccurate execution of the scheduling scheme, and even cause power grid failure or power outage, the present application provides a power grid intelligent scheduling method and system fusing RPA and AI.

[0007] The technical scheme provided by the embodiments of the present application is as follows:

[0008] First aspect:

[0009] The power grid intelligent scheduling method fusing RPA and AI provided by the embodiments of the present application is applied to a memory, comprising:

[0010] S1: Construct an artificial intelligence (AI) based power grid intelligent scheduling model, which specifically includes an energy use behavior prediction module, a power grid scheduling scheme creation module, and a scheduling scheme determination module.

[0011] S2: Use a three-layer stacked long short-term memory neural network to predict energy use behavior through the energy use behavior prediction module.

[0012] S3: Use a whale optimization algorithm to create a suboptimal power grid scheduling scheme based on the predicted energy use behavior through the power grid scheduling scheme creation module.

[0013] S4: Determine a target scheduling scheme from the suboptimal power grid scheduling scheme based on user habit deviation through the scheduling scheme determination module.

[0014] S5: Execute the target scheduling scheme through robot process automation (RPA) technology.

[0015] Second aspect:

[0016] The power grid intelligent scheduling system provided by the embodiment of the present application fuses RPA and AI, and includes:

[0017] A processor;

[0018] A memory having computer readable instructions stored thereon, which, when executed by the processor, implement the power grid intelligent scheduling method fusing RPA and AI as described in the first aspect.

[0019] Third aspect:

[0020] The computer readable storage medium provided by the embodiment of the present application has a computer program stored thereon, which, when executed by the processor, implements the power grid intelligent scheduling method fusing RPA and AI as described in the first aspect.

[0021] The technical scheme provided by the embodiment of the present application has at least the following beneficial effects:

[0022] (1) In the present application, after using a whale optimization algorithm to create a suboptimal power grid scheduling scheme, a target scheduling scheme is determined from the suboptimal power grid scheduling scheme based on user habit deviation, which focuses on considering user electricity use habits, so that users can obtain sufficient power supply to meet electricity use habits during high demand periods, ensuring daily life and work, and improving user satisfaction.

[0023] (2) In this invention, the target scheduling scheme is executed by robotic process automation (RPA) technology, which can automatically execute power grid scheduling tasks, significantly improve task processing speed and efficiency, reduce power grid scheduling errors caused by human negligence or mistakes, promote the intelligent and automated development of power grid scheduling, and improve the overall effect of power grid operation and user satisfaction. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 A flowchart illustrating a power grid intelligent dispatching method integrating RPA and AI, provided as an embodiment of the present invention;

[0026] Figure 2 This is a schematic diagram of the structure of a smart grid dispatching system that integrates RPA and AI, provided as an embodiment of the present invention. Detailed Implementation

[0027] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0028] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0029] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent.

[0030] In this embodiment of the invention, sometimes a subscript such as W1 may be mistakenly written as a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0031] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0032] Reference manual attached Figure 1 The diagram illustrates a flowchart of a smart grid scheduling method integrating RPA and AI provided by an embodiment of the present invention.

[0033] The present invention provides a power grid intelligent dispatching method integrating RPA and AI, the processing flow of which may include the following steps:

[0034] S1: Construct a smart grid dispatch model based on artificial intelligence (AI).

[0035] The smart grid dispatch model specifically includes: an energy use behavior prediction module, a grid dispatch scheme creation module, and a dispatch scheme determination module.

[0036] S2: Energy usage behavior is predicted using a three-layer stacked long short-term memory neural network through the energy usage behavior prediction module.

[0037] Among them, Long Short-Term Memory (LSTM) neural networks are a type of recurrent neural network (RNN) specifically designed for processing and predicting time series data. LSTM addresses the gradient vanishing and gradient exploding problems faced by traditional RNNs when processing long sequence data by introducing a special gating mechanism.

[0038] Optionally, the three-layer stacked long short-term memory neural network includes a first long short-term memory neural network, a second long short-term memory neural network, and a third long short-term memory neural network stacked together.

[0039] It should be noted that stacked LSTM networks, through multi-layer processing and feature extraction, can capture more complex and deeper time-series patterns. Each layer of the LSTM network can learn features at different levels, thereby improving the overall prediction accuracy and performance.

[0040] Optionally, the input layer of the first long short-term memory neural network is used to obtain the historical energy usage sequence. The hidden layer of the first long short-term memory neural network is connected to the input layer of the second long short-term memory neural network. The hidden layer of the second long short-term memory neural network is connected to the input layer of the third long short-term memory neural network. The output layer of the third long short-term memory neural network is connected to a fully connected layer. The fully connected layer is used to fuse the hidden states in the hidden layers of the first, second, and third long short-term memory neural networks to obtain a comprehensive hidden state.

[0041] It's worth noting that multi-layered LSTM structures can better capture short- and long-term dependencies in time series data. The first LSTM layer primarily learns short-term dependencies, the second LSTM layer further extracts medium-term dependencies, and the third LSTM layer captures longer-term dependencies. This multi-layered approach makes the model more adaptable to various temporal dependencies.

[0042] In one possible implementation, S2 specifically includes sub-steps S201 to S205:

[0043] S201: Input the historical energy usage sequence into the first long short-term neural network to obtain the first hidden state.

[0044] Specifically, for a certain layer of bidirectional long short-term neural network, the hidden state h is extracted as follows:

[0045] I t =Sigmoid(W XT X t +W HI h t-1 +b I )

[0046] F t =Sigmoid(W XF X t +W HF h t-1 +b F )

[0047] O t =Sigmoid(W XO X t +W HO h t-1 +b O )

[0048] C′ t =tanh(W XC X t +W HC h t-1 +b C )

[0049] C t =F t ·C t-1 +I t ·C t

[0050] h t =O t ·tanh(C t )

[0051] Among them, I tW represents the activation-output vector of the input gate at time t, Sigmoid() represents the Sigmoid activation function, and W... XI X represents the weight matrix between the segmented sequence and the input gate. t W represents the weighted word vector input at time t. HI b represents the weight matrix between the hidden state and the input gate. I F represents the bias term of the input gate. t W represents the activation and output vector of the gate forgotten at time t. XF W represents the weight matrix between the word segmentation sequence and the forget gate. HF b represents the weight matrix between the hidden state and the forget gate. F Indicates forgetting the bias term of the door, O t W represents the activation-output vector of the output gate at time t. XO W represents the weight matrix between the segmentation sequence and the output gate. HO b represents the weight matrix between the hidden state and the output gate. O C represents the bias term of the output gate. t Let C′ represent the activation output vector of the cell storage unit at time t. t C represents the candidate output vector of the cell storage unit at time t. t-1 W represents the activation output vector of the cell storage unit at time t-1, tanh() represents the tanh activation function, and W XC W represents the weight matrix between the segmented sequence and the cell storage unit. HC b represents the weight matrix between the hidden state and the cell storage unit. C h represents the bias term of the cell storage unit. t h represents the hidden state at time t. t-1 This represents the hidden state at time t-1.

[0052] S202: Input the first hidden state into the second long short-term neural network to obtain the second hidden state.

[0053] S203: Input the second hidden state into the third long short-term neural network to obtain the third hidden state.

[0054] It should be noted that the first long short-term neural network can capture short-term features and local patterns, the second long short-term neural network captures medium-term features and can integrate the output of the first layer to further extract useful information, and the third long short-term neural network can capture long-term features and integrate the information of the first two layers to perform deeper feature extraction.

[0055] S204: Merge the first hidden state, the second hidden state, and the third hidden state to obtain the combined hidden state:

[0056] Ht =ω1h 1t +ω2h 2t +ω3h 3t

[0057] Among them, H t Represents the integrated hidden state, h 1t Let h represent the first hidden state, ω1 represent the weight coefficient of the first hidden state, and h represent the weight coefficient of the first hidden state. 2t Let h represent the second hidden state, ω2 represent the weight coefficient of the second hidden state, and h represent the weight coefficient of the second hidden state. 3t ω3 represents the third hidden state, and ω3 represents the weight coefficient of the third hidden state.

[0058] Those skilled in the art can set the weight coefficients ω1, ω2, and ω3 of the first hidden state according to the actual situation, and the present invention does not limit them.

[0059] In this invention, the expressive power of the model is significantly enhanced by fusing hidden states at different levels. Each hidden state represents information at different time scales and feature levels. By combining these hidden states, the model can more accurately reflect the inherent structure and trends of the data, thereby improving its predictive performance.

[0060] S205: Predict energy use behavior based on the comprehensive hidden state:

[0061] Y = Sigmoid(W) h H t +b h )

[0062] Where Y represents the predicted energy consumption at future times, Sigmoid() represents the Sigmoid activation function, and W... h b represents the weight matrix between the fully connected layer and the prediction layer. h This represents the prediction layer bias term.

[0063] It should be noted that the energy consumption predicted using the Sigmoid activation function is a standardized value between [0,1], and inverse normalization is required to obtain the final predicted energy consumption value. Accordingly, the input historical energy consumption sequence needs to be normalized before input.

[0064] In this invention, by inputting the integrated hidden state into a fully connected layer and using the Sigmoid activation function to predict energy consumption, the rich feature information extracted by the multi-layer LSTM can be fully utilized to capture complex nonlinear relationships and generate standardized and stable prediction results.

[0065] S3: Through the power grid dispatch scheme creation module, based on predicted energy usage behavior, a suboptimal power grid dispatch scheme is created using the whale optimization algorithm.

[0066] In one possible implementation, S3 specifically includes sub-steps S301 to S304:

[0067] S301: Construct a power grid dispatch matrix to characterize the dispatch scheme.

[0068] Optionally, the power grid dispatch matrix is ​​specifically as follows:

[0069] Y t =D t E t

[0070]

[0071] Among them, Y t Let y represent the predicted energy consumption vector at time t. it This represents the predicted energy consumption of the i-th user at time t, where i = 1, 2, ..., m, and m represents the total number of users. t Let e ​​represent the energy supply prediction vector at time t. jt D represents the energy supply of the j-th power supply device at time t, where j = 1, 2, ..., n, and n represents the total number of power supply devices. t Let d represent the power grid dispatch matrix at time t. ijt This represents the amount of power supplied by the j-th power supply device to the j-th user at time t.

[0072] S302: Set power grid dispatch constraints.

[0073] Optionally, the power grid dispatch constraints specifically include: power balance constraints, equipment power constraints, and equipment capacity constraints.

[0074] Optionally, the power balance constraint is specifically as follows:

[0075]

[0076] in, e represents the amount of electricity purchased by the power grid at time t. jt This represents the energy supply of the j-th power supply device at time t, where j = 1, 2, ..., n, and n represents the total number of power supply devices. Let R represent the power flow at time t, and let I represent the load. t This represents the current at time t. This represents the load power at time t.

[0077] It should be noted that power balance constraints ensure that the total power generation of the power grid matches the total load at any given time, maintain power balance, avoid grid overload or insufficient power supply, and thus ensure the stable operation of the power grid.

[0078] Optionally, the device power constraint is specifically as follows:

[0079] P j,min ≤P j ≤P j,max

[0080] Among them, P j P represents the power of the j-th power supply device. j,min P represents the lower limit of the power supply of the j-th power supply device. j,max This represents the upper limit of the power supply of the j-th power supply device.

[0081] It should be noted that the equipment power constraint ensures that the power of each power supply device operates within its design range.

[0082] Optionally, the equipment capacity constraint is specifically as follows:

[0083] Q j,min ≤Q j ≤Q j,max

[0084] Among them, Q j Let Q represent the capacity of the j-th power supply device. j,min Q represents the lower limit of the capacity of the j-th power supply device. j,max This represents the upper limit of the capacity of the j-th power supply device.

[0085] It should be noted that the equipment capacity constraint ensures that the capacity of each power supply device operates within its design range.

[0086] S303: Under the constraints of power grid dispatching, construct the power grid dispatching objective function with the goals of reducing electricity costs, peak load ratio, and carbon emissions.

[0087] Optionally, the objective function for power grid dispatching is as follows:

[0088]

[0089] Wherein, min represents minimization, f represents the grid dispatch objective function, D represents the grid dispatch scheme, the grid dispatch scheme is composed of the grid dispatch matrix at each time point, C represents electricity cost, A represents peak load ratio, B represents carbon emissions, t = 1, 2, ..., T, T represents the total observation time, β1 represents the weighting coefficient of electricity cost, β2 represents the weighting coefficient of peak load ratio, and β3 represents the weighting coefficient of carbon emissions.

[0090] Those skilled in the art can set the weighting coefficients β1 for electricity cost, β2 for peak load ratio, and β3 for carbon emissions according to actual conditions; this invention does not impose any limitations.

[0091] In this invention, by constructing a comprehensive optimization objective function and optimizing the scheduling scheme based on it, it is possible to effectively reduce electricity costs, balance loads, reduce carbon emissions, and improve the economy, reliability, and environmental friendliness of the power grid.

[0092] Optionally, the electricity cost is as follows:

[0093]

[0094] in, This represents the electricity price at time t. λ represents the amount of electricity purchased by the power grid at time t. j Let e ​​represent the unit power generation cost of the j-th power supply device. jt Let represent the energy supply of the j-th power supply device at time t, where j = 1, 2, ..., n, and n represents the total number of power supply devices.

[0095] Optionally, the peak load ratio is specifically:

[0096]

[0097] Where max represents the maximum value, d it d represents the amount of power supplied to the i-th user at time t. ijt Let represent the amount of power supplied by the j-th power supply device to the i-th user at time t, where i = 1, 2, ..., m, m represents the total number of users, and j = 1, 2, ..., n, n represents the total number of power supply devices.

[0098] It should be noted that the peak load ratio reflects the degree of load balance. By optimizing the objective function, the load curve can be smoothed, peak load can be reduced, the pressure on the power grid during peak hours can be alleviated, the service life of power grid equipment can be extended, and the operational stability of the power grid can be improved.

[0099] Optionally, the carbon emissions are as follows:

[0100]

[0101] Where, γ j Let e ​​represent the carbon emission coefficient of the j-th power supply device. jt Let represent the energy supply of the j-th power supply device at time t, where j = 1, 2, ..., n, and n represents the total number of power supply devices.

[0102] S304: Based on the power grid dispatch objective function, use the whale optimization algorithm to create a suboptimal power grid dispatch scheme.

[0103] Among them, the Whale Optimization Algorithm (WOA) is an optimization algorithm based on the foraging behavior of whales in nature. It simulates the bubble net foraging behavior of humpback whales in order to search for the global optimum.

[0104] The reciprocal of the power grid scheduling objective function is used as the fitness function of the whale optimization algorithm.

[0105] Initialize a whale population, which contains multiple individual whales. The position of each individual whale in the search space represents a feasible power grid scheduling scheme.

[0106] Generate a random number r1 between 0 and 1, and calculate the first coefficient A:

[0107] A = 2ar1-a

[0108] Where a represents the adaptive convergence factor.

[0109] Optionally, the adaptive convergence factor a is specifically:

[0110]

[0111] Where t represents the current iteration number and T represents the maximum iteration number.

[0112] In this invention, the search range is dynamically adjusted at different stages of the algorithm through the design of an adaptive convergence factor 'a'. In the early stages, the search range is larger, which is beneficial for global search; in the later stages, the search range gradually shrinks, which is beneficial for fine-grained local search. This dynamic adjustment mechanism effectively balances global and local search.

[0113] If |A| < 1, proceed to the next step. If |A| ≥ 1, enter the random search mechanism.

[0114] Generate a random variable p.

[0115] When p < 0.5, the contraction and encirclement mechanism is activated. When p ≥ 0.5, the spiral mechanism is activated.

[0116] Once the current global optimal position is determined, other individual whales move towards that position, updating their own positions through a shrinking encirclement mechanism.

[0117] X(t+1)=X best (t)-A·D

[0118] D = |C·X best (t)-X(t)|

[0119] C = 2r1

[0120] Where X(t+1) represents the position of the individual whale at the (t+1)th iteration, X best X(t) represents the global optimal position at the t-th iteration, A represents the first coefficient, D represents the randomized distance between the individual whale position and the current global optimal position, C represents the second coefficient, t represents the current iteration number, and X(t) represents the individual whale position at the t-th iteration.

[0121] In this invention, updating the position of individual whales through a shrinking encirclement mechanism can effectively simulate natural behavior, balance exploration and development, improve convergence speed, and enhance the diversity of solutions and the robustness of the algorithm.

[0122] Generate a random number r2 between 0 and 1. The whale updates its position by spiraling upwards and continuously shrinking the encirclement.

[0123]

[0124] D * =|X best (t)-X(t)|

[0125] Among them, D * The distance between the individual whale's position and the current global best position is represented by , e represents the natural constant, b represents the logarithmic spiral constant, and l represents a random number between -1 and 1.

[0126] In this invention, different spiral paths (cosine or sine) are selected based on the value of the random number r2 to increase path diversity. By moving upwards in a spiral motion and continuously shrinking the enclosing circle, the whale optimization algorithm can effectively balance global search and local optimization, improve the diversity and flexibility of search paths, enhance the robustness and adaptability of the algorithm, quickly approach the global optimum, and ultimately improve optimization effect and efficiency. This method combines the spiral mechanism and the shrinking enclosing mechanism, and further enhances the comprehensiveness and effectiveness of the search by randomly selecting the movement mode.

[0127] Based on the positions of individual whales relative to each other, their own positions are updated through a random search mechanism:

[0128] X(t+1)=X rand -A·D rand

[0129] D rand =|C·X rand -X(t)|

[0130] Among them, X rand This indicates the location of a randomly selected individual whale.

[0131] In this invention, by randomly selecting the locations of other individual whales for updating, the algorithm can escape the current search area and explore a wider space. This effectively avoids the algorithm getting stuck in local optima and enhances its global search capability.

[0132] Update the global optimal position.

[0133] Apply Cauchy mutation perturbation to the global optimal position:

[0134]

[0135] in, represents the global optimal position after Cauchy mutation perturbation, and cauchy(0,1) represents the Cauchy operator.

[0136] In this invention, the Cauchy distribution exhibits heavy-tailed characteristics, which can generate significant perturbations. This allows the algorithm to perform a wide-ranging search around the global optimum, helping to escape local optima and enhancing global search capabilities.

[0137] Determine if the current iteration count has reached the maximum iteration count. If so, output the power grid scheduling scheme represented by the preset number of whale individuals with the highest fitness as the suboptimal power grid scheduling scheme. Otherwise, return to continue iterating.

[0138] Those skilled in the art can set the preset number according to the actual situation, and the present invention does not limit it. Optionally, the preset number is 10, that is, the power grid dispatching schemes represented by the 10 whale individuals with the highest fitness ranking are determined as the suboptimal power grid dispatching schemes.

[0139] S4: The scheduling scheme determination module determines the target scheduling scheme from the suboptimal power grid scheduling schemes based on the user habit deviation degree.

[0140] In one possible implementation, S4 specifically includes sub-steps S401 and S402:

[0141] S401: Calculate the user habit deviation for each suboptimal power grid dispatch scheme:

[0142]

[0143] Where σ represents the degree of user habit deviation, u i This represents the electricity consumption habits of the i-th user at time t, where i = 1, 2, ..., m, m represents the total number of users, and d represents the electricity consumption habits of the i-th user at time t. it Let t represent the amount of power supplied to the i-th user at time t, where t = 1, 2, ..., T, and T represents the total observation duration.

[0144] S402: Select the suboptimal power grid dispatch scheme with the smallest user habit deviation as the target dispatch scheme.

[0145] In this invention, by calculating the user habit deviation degree, the scheduling scheme that best matches the user's actual electricity consumption habits can be selected, thereby maximizing the satisfaction of the user's electricity needs and improving user satisfaction.

[0146] S5: Executes target scheduling schemes through Robotic Process Automation (RPA) technology.

[0147] Robotic Process Automation (RPA) is a technology that uses software robots (or "digital employees") to automatically execute rule-based repetitive business processes.

[0148] Specifically, the software robot sends instructions to the control system according to the scheduling plan, adjusting the output power of the main generator sets and starting the standby generator sets. Simultaneously, the robot monitors the power grid's operating status in real time, checking the generator set operation to ensure the correct execution of scheduling instructions. During the scheduling process, the robot generates a scheduling execution report, recording the operations and results of each step, and sends it to relevant personnel or systems for archiving and analysis.

[0149] The beneficial effects of the technical solutions provided by the embodiments of the present invention include at least the following:

[0150] (1) In this invention, after creating a suboptimal power grid dispatch scheme using the whale optimization algorithm, a target dispatch scheme is determined from the suboptimal power grid dispatch scheme based on the user habit deviation degree. This takes into account the user's electricity consumption habits, so that users can obtain sufficient power supply during high demand periods to meet their electricity consumption habits, ensure daily life and work, and improve user satisfaction.

[0151] (2) In this invention, the target scheduling scheme is executed by robotic process automation (RPA) technology, which can automatically execute power grid scheduling tasks, significantly improve task processing speed and efficiency, reduce power grid scheduling errors caused by human negligence or mistakes, promote the intelligent and automated development of power grid scheduling, and improve the overall effect of power grid operation and user satisfaction.

[0152] Reference manual attached Figure 2 The diagram shows a structural schematic of a smart grid dispatching system that integrates RPA and AI provided by the present invention.

[0153] The present invention also provides a power grid intelligent dispatching system 20 integrating RPA and AI, applied to the above-mentioned power grid intelligent dispatching method integrating RPA and AI, comprising:

[0154] Processor 201;

[0155] The memory 202 stores computer-readable instructions, which, when executed by the processor 201, implement the intelligent power grid dispatching method integrating RPA and AI as described in the method embodiment.

[0156] The power grid intelligent dispatching system 20 integrating RPA and AI provided by the present invention can execute the above-mentioned power grid intelligent dispatching method integrating RPA and AI and achieve the same or similar technical effects. To avoid duplication, the present invention will not elaborate further.

[0157] The beneficial effects of the technical solutions provided by the embodiments of the present invention include at least the following:

[0158] (1) In this invention, after creating a suboptimal power grid dispatch scheme using the whale optimization algorithm, a target dispatch scheme is determined from the suboptimal power grid dispatch scheme based on the user habit deviation degree. This takes into account the user's electricity consumption habits, so that users can obtain sufficient power supply during high demand periods to meet their electricity consumption habits, ensure daily life and work, and improve user satisfaction.

[0159] (2) In this invention, the target scheduling scheme is executed by robotic process automation (RPA) technology, which can automatically execute power grid scheduling tasks, significantly improve task processing speed and efficiency, reduce power grid scheduling errors caused by human negligence or mistakes, promote the intelligent and automated development of power grid scheduling, and improve the overall effect of power grid operation and user satisfaction.

[0160] It should be understood that the processor in the embodiments of the present invention can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0161] It should also be understood that the memory in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0162] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0163] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0164] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0165] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0166] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software 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 implementations should not be considered beyond the scope of this invention.

[0167] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0168] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0169] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0170] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0171] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0172] This invention provides a computer-readable storage medium storing a computer program thereon, characterized in that, when the program is executed by a processor, it implements the intelligent power grid dispatching method integrating RPA and AI as described in the method embodiment.

[0173] The present invention provides a computer-readable storage medium that can implement the steps and effects of the above-described method embodiment of the power grid intelligent dispatching method integrating RPA and AI. To avoid repetition, the present invention will not repeat them.

[0174] The beneficial effects of the technical solutions provided by the embodiments of the present invention include at least the following:

[0175] (1) In this invention, after creating a suboptimal power grid dispatch scheme using the whale optimization algorithm, a target dispatch scheme is determined from the suboptimal power grid dispatch scheme based on the user habit deviation degree. This takes into account the user's electricity consumption habits, so that users can obtain sufficient power supply during high demand periods to meet their electricity consumption habits, ensure daily life and work, and improve user satisfaction.

[0176] (2) In this invention, the target scheduling scheme is executed by robotic process automation (RPA) technology, which can automatically execute power grid scheduling tasks, significantly improve task processing speed and efficiency, reduce power grid scheduling errors caused by human negligence or mistakes, promote the intelligent and automated development of power grid scheduling, and improve the overall effect of power grid operation and user satisfaction.

[0177] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0178] The following points need to be explained: (1) The accompanying drawings of the embodiments of the present invention only involve the structures involved in the embodiments of the present invention, and other structures can refer to the general design.

[0179] (2) For clarity, the thickness of layers or regions is enlarged or reduced in the drawings used to describe embodiments of the present invention; that is, these drawings are not drawn to actual scale. It is understood that when an element such as a layer, film, region, or substrate is referred to as being “above” or “below” another element, the element may be “directly” located “above” or “below” the other element, or there may be intermediate elements.

[0180] (3) Where there is no conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other to obtain new embodiments.

[0181] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. The scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A smart grid dispatching method integrating RPA and AI, characterized in that, Applied to memory, including: S1: Construct a smart grid dispatching model based on artificial intelligence (AI). The smart grid dispatching model specifically includes: an energy use behavior prediction module, a grid dispatching scheme creation module, and a dispatching scheme determination module. S2: The energy usage behavior prediction module uses a three-layer stacked long short-term memory neural network to predict energy usage behavior. S3: Through the power grid dispatching scheme creation module, based on the predicted energy usage behavior, a suboptimal power grid dispatching scheme is created using the whale optimization algorithm; S4: The target scheduling scheme is determined from the suboptimal power grid scheduling schemes based on the user habit deviation degree through the scheduling scheme determination module; S5: Execute the target scheduling scheme using Robotic Process Automation (RPA) technology.

2. The intelligent power grid dispatching method integrating RPA and AI according to claim 1, characterized in that, The three-layer stacked long short-term memory neural network includes a first long short-term memory neural network, a second long short-term memory neural network, and a third long short-term memory neural network stacked together. The input layer of the first long short-term memory neural network is used to obtain historical energy usage sequences. The hidden layer of the first long short-term memory neural network is connected to the input layer of the second long short-term memory neural network. The hidden layer of the second long short-term memory neural network is connected to the input layer of the third long short-term memory neural network. The output layer of the third long short-term memory neural network is connected to a fully connected layer. The fully connected layer is used to fuse the hidden states in the hidden layers of the first long short-term memory neural network, the second long short-term memory neural network, and the third long short-term memory neural network to obtain a comprehensive hidden state.

3. The intelligent power grid dispatching method integrating RPA and AI according to claim 2, characterized in that, S2 specifically includes: S201: Input the historical energy usage sequence into the first long short-term neural network to obtain the first hidden state; S202: Input the first hidden state into the second long short-term neural network to obtain the second hidden state; S203: Input the second hidden state into the third long short-term neural network to obtain the third hidden state; S204: The first hidden state, the second hidden state, and the third hidden state are fused to obtain a combined hidden state: H t =ω1h 1t +ω2h 2t +ω3h 3t Among them, H t Represents the integrated hidden state, h 1t Let h represent the first hidden state, ω1 represent the weight coefficient of the first hidden state, and h represent the weight coefficient of the first hidden state. 2t Let h represent the second hidden state, ω2 represent the weight coefficient of the second hidden state, and h represent the weight coefficient of the second hidden state. 3t ω3 represents the weight coefficient of the third hidden state; S205: Based on the aforementioned integrated hidden state, predict energy usage behavior: Y=Sigmoid(W h H t +b h ) Where Y represents the predicted energy consumption at future times, Sigmoid() represents the Sigmoid activation function, and W... h b represents the weight matrix between the fully connected layer and the prediction layer. h This represents the prediction layer bias term.

4. The intelligent power grid dispatching method integrating RPA and AI according to claim 1, characterized in that, S3 specifically includes: S301: Construct a power grid dispatch matrix to characterize the dispatch scheme; S302: Set power grid dispatch constraints; S303: Under the constraints of the power grid dispatching conditions, construct a power grid dispatching objective function with the goals of reducing electricity costs, peak load ratio, and carbon emissions; S304: Based on the power grid scheduling objective function, use the whale optimization algorithm to create a suboptimal power grid scheduling scheme.

5. The intelligent power grid dispatching method integrating RPA and AI according to claim 4, characterized in that, The power grid dispatch matrix is ​​specifically as follows: Y t =D t E t Among them, Y t Let y represent the predicted energy consumption vector at time t. it This represents the predicted energy consumption of the i-th user at time t, where i = 1, 2, ..., m, and m represents the total number of users. t Let e ​​represent the energy supply prediction vector at time t. jt D represents the energy supply of the j-th power supply device at time t, where j = 1, 2, ..., n, and n represents the total number of power supply devices. t Let d represent the power grid dispatch matrix at time t. ijt This represents the amount of power supplied by the j-th power supply device to the j-th user at time t.

6. The intelligent power grid dispatching method integrating RPA and AI according to claim 4, characterized in that, The power grid dispatch constraints specifically include: power balance constraints, equipment power constraints, and equipment capacity constraints.

7. The intelligent power grid dispatching method integrating RPA and AI according to claim 4, characterized in that, The specific objective function for power grid scheduling is as follows: Wherein, min represents minimization, f represents the grid dispatch objective function, D represents the grid dispatch scheme, the grid dispatch scheme is composed of the grid dispatch matrix at each time point, C represents electricity cost, A represents peak load ratio, B represents carbon emissions, t = 1, 2, ..., T, T represents the total observation time, β1 represents the weighting coefficient of electricity cost, β2 represents the weighting coefficient of peak load ratio, and β3 represents the weighting coefficient of carbon emissions.

8. The intelligent power grid dispatching method integrating RPA and AI according to claim 7, characterized in that, The electricity cost is specifically as follows: in, This represents the electricity price at time t. λ represents the amount of electricity purchased by the power grid at time t. j Let e ​​represent the unit power generation cost of the j-th power supply device. jt Let j represent the energy supply of the j-th power supply device at time t, where j = 1, 2, ..., n, and n represents the total number of power supply devices. The peak load ratio is specifically: Where max represents the maximum value, d it d represents the amount of power supplied to the i-th user at time t. ijt Let represent the amount of power supplied by the j-th power supply device to the i-th user at time t, where i = 1, 2, ..., m, m represents the total number of users, and j = 1, 2, ..., n, n represents the total number of power supply devices; The carbon emissions are specifically: Where, γ j Let e ​​represent the carbon emission coefficient of the j-th power supply device. jt Let represent the energy supply of the j-th power supply device at time t, where j = 1, 2, ..., n, and n represents the total number of power supply devices.

9. The intelligent power grid dispatching method integrating RPA and AI according to claim 1, characterized in that, S4 specifically includes: S401: Calculate the user habit deviation for each suboptimal power grid dispatch scheme: Where σ represents the degree of user habit deviation, u i This represents the electricity consumption habits of the i-th user at time t, where i = 1, 2, ..., m, m represents the total number of users, and d represents the electricity consumption habits of the i-th user at time t. it Let t represent the amount of power supplied to the i-th user at time t, where t = 1, 2, ..., T, and T represents the total observation duration. S402: Select the suboptimal power grid dispatch scheme with the smallest user habit deviation as the target dispatch scheme.

10. A smart grid dispatching system integrating RPA and AI, characterized in that, include: processor; A memory storing computer-readable instructions, which, when executed by the processor, implement the power grid intelligent dispatching method integrating RPA and AI as described in any one of claims 1 to 9.

Images