Line load prediction method and device, electronic equipment and storage medium

Abstract

The invention discloses a line load prediction method and device, electronic equipment and a storage medium. The method comprises the following steps: determining a real-time prediction date of a load prediction line and a prediction reference date corresponding to the real-time prediction date, and determining a real-time prediction moment in the real-time prediction date; determining a plurality of first reference moments in the real-time prediction date and a plurality of second reference moments in the real-time reference date according to the real-time prediction moment; determining first load data corresponding to each first reference moment and second load data corresponding to each second reference moment, and determining weather date data corresponding to the real-time prediction date; and determining a load prediction result of the load prediction line at a real-time prediction moment of the real-time prediction date according to the first load data corresponding to each first reference moment, the second load data corresponding to each second reference moment and the weather date data. According to the technical scheme, the accuracy of line load prediction can be improved.

Description

Technical Field

[0001] The present invention relates to the field of computer application technology, and in particular to a line load prediction method, device, electronic equipment and storage medium. Background Art

[0002] With the development of society, people's daily electricity consumption has increased year by year. Correspondingly, the number of newly built lines and newly invested substations has also increased year by year. Therefore, the workload of safety monitoring of the load conditions of distribution lines has become increasingly heavy.

[0003] In related technologies, the distribution lines are usually monitored for safety by automatically predicting the load conditions of the lines through traditional neural network models to reduce the difficulty of the staff's work. However, prediction errors occasionally occur, and as the prediction task becomes more demanding, the prediction errors become more and more obvious. Summary of the invention

[0004] The present invention provides a line load prediction method, device, electronic device and storage medium to solve the problem that the accuracy of line load prediction using traditional neural network models in related technologies is low and it is difficult to meet the prediction accuracy requirements of users under a large number of prediction tasks.

[0005] According to one aspect of the present invention, a line load prediction method is provided, wherein the method comprises:

[0006] Determine a real-time prediction date of a load prediction line and a prediction reference date corresponding to the real-time prediction date, and determine a real-time prediction moment in the real-time prediction date;

[0007] Determine a plurality of first reference moments in the real-time prediction date and a plurality of second reference moments in the real-time reference date according to the real-time prediction moment;

[0008] Determine the first load data corresponding to each of the first reference moments and the second load data corresponding to each of the second reference moments, and determine the weather date data corresponding to the real-time forecast date;

[0009] Determine the load forecast result of the load forecast line at the real-time forecast time on the real-time forecast date according to the first load data corresponding to each of the first reference times, the second load data corresponding to each of the second reference times, and the weather date data.

[0010] According to another aspect of the present invention, a line load prediction device is provided, wherein the device comprises:

[0011] A date and time determination module, used to determine the real-time prediction date of the load prediction line and the prediction reference date corresponding to the real-time prediction date, and determine the real-time prediction time in the real-time prediction date;

[0012] A reference time determination module, used to determine a plurality of first reference times in the real-time prediction date and a plurality of second reference times in the real-time reference date according to the real-time prediction time;

[0013] a characteristic data determination module, used to determine the first load data corresponding to each of the first reference moments and the second load data corresponding to each of the second reference moments, and to determine the weather date data corresponding to the real-time forecast date;

[0014] The line load prediction module is used to determine the load prediction result of the load prediction line at the real-time prediction moment of the real-time prediction date based on the first load data corresponding to each first reference moment, the second load data corresponding to each second reference moment and the weather date data.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] at least one processor; and

[0017] a memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the line load prediction method described in any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the line load prediction method described in any embodiment of the present invention when executed.

[0020] The technical solution of the embodiment of the present invention determines the real-time prediction date of the load prediction line and the prediction reference date corresponding to the real-time prediction date, determines the real-time prediction moment in the real-time prediction date; determines multiple first reference moments in the real-time prediction date and multiple second reference moments in the real-time reference date according to the real-time prediction moment; determines the first load data corresponding to each of the first reference moments and the second load data corresponding to each of the second reference moments, and determines the weather date data corresponding to the real-time prediction date; determines the load prediction result of the load prediction line at the real-time prediction moment of the real-time prediction date according to the first load data corresponding to each of the first reference moments, the second load data corresponding to each of the second reference moments, and the weather date data. The technical effect of improving the accuracy of line load prediction is achieved.

[0021] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 is a flow chart of a line load prediction method provided according to Embodiment 1 of the present invention;

[0024] Figure 2 is a flow chart of a line load prediction method provided according to Embodiment 2 of the present invention;

[0025] Figure 3 is an overall flow chart of a line load prediction method provided according to an embodiment of the present invention;

[0026] Figure 4 is a structural diagram of a back propagation neural network provided according to an embodiment of the present invention;

[0027] Figure 5 is a flow chart of a load forecasting model obtained by training according to an embodiment of the present invention;

[0028] Figure 6 is a structural schematic diagram of a line load prediction device provided according to Embodiment 3 of the present invention;

[0029] Figure 7It is a structural schematic diagram of an electronic device for implementing the line load prediction method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme 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 only 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 should fall within the scope of protection of the present invention.

[0031] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0032] Embodiment 1

[0033] Figure 1 A flowchart of a line load prediction method is provided for the first embodiment of the present invention. This embodiment is applicable to the case where line load prediction is performed through a deep learning model. The method can be performed by a line load prediction device. The line load prediction device can be implemented in the form of hardware and / or software. The line load prediction device can be configured in a computer. Figure 1 As shown, the method includes:

[0034] S110, determining a real-time prediction date of a load prediction line and a prediction reference date corresponding to the real-time prediction date, and determining a real-time prediction moment in the real-time prediction date.

[0035] The load prediction line can be understood as a target line for load prediction. The load prediction line can be a line in a target substation.

[0036] The real-time forecast date may be understood as the real-time date for performing load forecasting, or may be understood as the target date for performing load forecasting.

[0037] The prediction reference date can be understood as a date that needs to be referenced for load prediction. In an embodiment of the present invention, the prediction reference date can be a historical date adjacent to the prediction reference date.

[0038] The real-time prediction time may be understood as the real-time time for performing the review prediction, or may be understood as the target time for performing the review prediction.

[0039] In the embodiment of the present invention, the real-time prediction time may be a future time relative to the current time. That is, the present invention predicts the line load condition at a future time to avoid line accidents caused by line overload.

[0040] S120. Determine a plurality of first reference moments in the real-time prediction date and a plurality of second reference moments in the real-time reference date according to the real-time prediction moment.

[0041] The first reference time and the second reference time can be understood as the time required for reference in load forecasting at the real-time forecasting time. In an embodiment of the present invention, the plurality of second reference times in the real-time reference date correspond to the first reference time and the real-time forecasting time in the real-time forecasting date.

[0042] Exemplarily, the real-time prediction date is A year A month A day, and the prediction reference date is A year A month (A-1) day. When the frequency of real-time load prediction is 1 time / h, if the real-time prediction moment is the nth moment (n is 1-24) in the prediction reference date, then the first reference moment may include the n-1th moment and the n-2th moment in the real-time prediction date; correspondingly, the second reference moment may include the nth moment, the n-1th moment and the n-2th moment in the prediction reference date. Alternatively, if the first reference moment may include the n-1th moment, the n-2th moment and the n-3th moment in the real-time prediction date, correspondingly, the second reference moment may include the nth moment, the n-1th moment, the n-2th moment and the n-3th moment in the prediction reference date, wherein, when the calculated reference moment is a negative value, it may be pushed forward to the adjacent previous historical date.

[0043] S130. Determine the first load data corresponding to each of the first reference moments and the second load data corresponding to each of the second reference moments, and determine the weather date data corresponding to the real-time forecast date.

[0044] The first load data may be understood as the load data at the first reference time, and the second load data may be understood as the load data at the second reference time.

[0045] Optionally, the load data may be one or more data values ​​that can characterize the load characteristics of the load forecasting line. In the embodiment of the present invention, the selection of the load data may be preset according to the scenario requirements, and is not specifically limited here.

[0046] The weather date data may be data representing weather characteristics and date characteristics. In the embodiment of the present invention, the selection of the weather date data may be preset according to scene requirements, and is not specifically limited here.

[0047] Optionally, the weather date data may include weather type, first temperature, second temperature and date type.

[0048] The weather type may be understood as the type of weather on the real-time forecast date. Optionally, the weather type may include a sunny type, a cloudy type, a light rain type, a moderate rain type, or a heavy rain type.

[0049] The first temperature and the second temperature may be two extreme temperatures in the real-time forecast date. Optionally, the first temperature and the second temperature may be the highest temperature and the lowest temperature, respectively.

[0050] In the embodiment of the present invention, the method for determining the weather type, the first temperature, and the second temperature is not limited in the embodiment of the present invention. For example, the weather data such as the weather type, the first temperature, and the second temperature are obtained by using weather forecasting related technology.

[0051] The date type may be understood as the type of the real-time predicted date. Optionally, the date type may include a working day and / or a non-working day.

[0052] S140. Determine the load forecast result of the load forecast line at the real-time forecast moment of the real-time forecast date based on the first load data corresponding to each of the first reference moments, the second load data corresponding to each of the second reference moments, and the weather date data.

[0053] In the embodiment of the present invention, the load data corresponding to different reference times may be different or the same.

[0054] The load prediction result may be understood as a result of line load prediction. Optionally, the load prediction result may include line overload and / or line non-overload.

[0055] Optionally, the load forecast result includes line overload and / or line non-overload, and after determining the load forecast result of the load forecast line at the real-time forecast time on the real-time forecast date, the method further includes:

[0056] When the load forecast result indicates that the line is overloaded, a line overload warning is issued through a target warning device.

[0057] The target warning device can be understood as a warning device. In the embodiment of the present invention, the target warning device can be preset according to the scene requirements, and is not specifically limited here.

[0058] Based on the technical solution of the above embodiment, an early warning is implemented before the line may be overloaded, thereby avoiding line accidents that may be caused by line overload.

[0059] The technical solution of the embodiment of the present invention determines the real-time prediction date of the load prediction line and the prediction reference date corresponding to the real-time prediction date, determines the real-time prediction moment in the real-time prediction date; determines multiple first reference moments in the real-time prediction date and multiple second reference moments in the real-time reference date according to the real-time prediction moment; determines the first load data corresponding to each of the first reference moments and the second load data corresponding to each of the second reference moments, and determines the weather date data corresponding to the real-time prediction date; determines the load prediction result of the load prediction line at the real-time prediction moment of the real-time prediction date according to the first load data corresponding to each of the first reference moments, the second load data corresponding to each of the second reference moments, and the weather date data. The technical effect of improving the accuracy of line load prediction is achieved.

[0060] Embodiment 2

[0061] Figure 2 This is a flow chart of a line load forecasting method provided in the second embodiment of the present invention. This embodiment is to refine the load forecast result of the load forecast line at the real-time forecast time of the real-time forecast date according to the first load data corresponding to each first reference time, the second load data corresponding to each second reference time and the weather date data described in the above embodiment. Figure 2 As shown, the method includes:

[0062] S210. Determine a real-time prediction date of a load prediction line and a prediction reference date corresponding to the real-time prediction date, and determine a real-time prediction moment in the real-time prediction date.

[0063] S220. Determine a plurality of first reference moments in the real-time prediction date and a plurality of second reference moments in the real-time reference date according to the real-time prediction moment.

[0064] S230. Determine the first load data corresponding to each of the first reference moments and the second load data corresponding to each of the second reference moments, and determine the weather date data corresponding to the real-time forecast date.

[0065] S240. Perform load forecasting on the first load data corresponding to each of the first reference moments, the second load data corresponding to each of the second reference moments, the weather type, the first temperature, the second temperature and the date type through a load forecasting model, and determine the load forecast result of the load forecasting line at the real-time forecast moment on the real-time forecast date.

[0066] In an embodiment of the present invention, the load forecasting model is obtained by training a back propagation neural network based on sample data.

[0067] The load forecasting model can be understood as a model with load forecasting function. In the embodiment of the present invention, the input data of the load forecasting model are load data corresponding to multiple reference times and weather date data corresponding to the real-time forecast date.

[0068] The back propagation neural network (BPNN) can be understood as a multi-layer feedforward neural network.

[0069] Optionally, the sample data includes training samples and test samples; before performing load forecasting on the first load data corresponding to each first reference moment, the second load data corresponding to each second reference moment, the weather type, the first temperature, the second temperature and the date type input by the load forecasting model, the further comprising:

[0070] Determine the number of input nodes and the number of output nodes of the back propagation neural network, determine multiple candidate numbers of hidden layer nodes according to the number of input nodes and the number of output nodes, and construct the back propagation neural network corresponding to each candidate number of hidden layer nodes;

[0071] For each of the back propagation neural networks corresponding to the candidate hidden layer node number, the number of input nodes, the number of output nodes and the transfer function of the hidden layer of the back propagation neural network are set, and the candidate prediction model is obtained by training the back propagation neural network through the training samples;

[0072] For each of the candidate prediction models, testing the candidate prediction model using the test sample to obtain a model test result;

[0073] The load forecasting model among the candidate forecasting models is determined based on the model testing result.

[0074] The number of input nodes can be understood as the number of input data nodes of the back propagation neural network in the structure. The number of output nodes can be understood as the number of output data nodes of the back propagation neural network in the structure. In the embodiment of the present invention, the number of input nodes and the number of output nodes can be preset according to user needs and / or scenario needs, and are not specifically limited here.

[0075] Optionally, determining the number of input nodes of the back propagation neural network includes:

[0076] The number of input nodes of the back propagation neural network is determined according to the preset reference time quantity information and the preset characteristic data quantity information related to the weather date.

[0077] Exemplarily, user requirements and / or scenario requirements are: the first reference moment includes the n-1th moment and the n-2th moment in the real-time prediction date, the second reference moment may include the nth moment, the n-1th moment and the n-2th moment in the prediction reference date, and the weather date data includes weather type, first temperature, second temperature and date type; the corresponding number of input nodes is 9.

[0078] Alternatively, the user requirements and / or scenario requirements are: the first reference time may include the n-1th moment, the n-2th moment and the n-3th moment in the real-time prediction date, the second reference time may include the nth moment, the n-1th moment, the n-2th moment and the n-3th moment in the prediction reference date, and the weather date data includes weather type, first temperature, second temperature and date type; then the corresponding number of input nodes is 11.

[0079] In the embodiment of the present invention, optionally, the number of the output nodes may be 1. The output node is used to output the load forecasting result.

[0080] The number of candidate hidden layer nodes can be understood as the number of nodes in the hidden layer of the back propagation neural network in structure. In an embodiment of the present invention, the number of candidate hidden layer nodes can be determined according to the number of input nodes and the number of output nodes.

[0081] The candidate prediction model can be understood as a candidate prediction model. In the embodiment of the present invention, for each of the candidate hidden layer nodes of the back propagation neural network, a corresponding candidate prediction model can be trained.

[0082] The model test result may be a test result that can characterize the line load prediction accuracy of the candidate prediction model. Optionally, the model test result may be the accuracy rate of the line load prediction of the candidate prediction model.

[0083] Specifically, the determining of the number of candidate hidden layer nodes according to the number of input nodes and the number of output nodes can be implemented based on the following formula:

[0084] s = log2m;

[0085]

[0086] s=sqrt(0.43mn+0.12nn+2.54m+0.77n+0.35)+0.51;

[0087] s≤m-1;

[0088] Among them, s represents the number of candidate hidden layer nodes; m represents the number of input nodes; n represents the number of output nodes; α represents a constant between 0 and 10, and the selection of α can be preset according to scene requirements and is not specifically limited here.

[0089] In the embodiment of the present invention, multiple numbers of candidate hidden layer nodes can be obtained based on the above formula.

[0090] In the embodiment of the present invention, the training sample may include training data and training labels, and the test sample may include test data and test labels, wherein the label may represent the actual load condition of the sample.

[0091] Optionally, the setting of the transfer function of the hidden layer of the back propagation neural network includes:

[0092] The neural transfer function is set as the transfer function of the hidden layer in the back-propagation neural network.

[0093] It is understandable that there are many common transfer functions of hidden layers.

[0094] In the embodiment of the present invention, a neural transfer function, that is, a tansig() type transfer function is used as a transfer function of a hidden layer in the back propagation neural network.

[0095] The technical solution of the embodiment of the present invention performs load forecasting on the first load data corresponding to each of the first reference moments, the second load data corresponding to each of the second reference moments, the weather type, the first temperature, the second temperature and the date type through a load forecasting model, and determines the load forecast result of the load forecasting line at the real-time forecasting moment of the real-time forecasting date, wherein the load forecasting model is obtained by training a back propagation neural network based on sample data. The structure of the back propagation neural network adopted in the present invention is suitable for the application scenario of the present invention of predicting one-dimensional target data based on multi-dimensional reference data, and the load forecasting model obtained by back propagation neural network training is used to predict line load, which can effectively ensure the accuracy of line load prediction.

[0096] The above embodiment is further described as follows:

[0097] Figure 3 It is an overall flow chart of a line load prediction method provided according to an embodiment of the present invention.

[0098] like Figure 3 As shown, the overall process of the line load prediction method can be:

[0099] 1. Determine the number of input nodes, the number of output nodes, the number of candidate hidden layer nodes, and the transfer function, and determine multiple BPNNs corresponding to the above data.

[0100] 2. For each BPNN, the data of the input nodes of the input layer (including: sample load data, weather type, weather temperature, date type, etc.) are normalized and preprocessed.

[0101] 3. Select a certain amount of sample data to train the BPNN, and improve the weight threshold matrix of the neural network from the last layer forward through the error between the output result and the true result (sample label), and obtain the desired neural network (the model test result) after the error is reduced to the set value or a certain number of iterations.

[0102] 4. Use the model test results to perform load forecasting on the load data corresponding to multiple reference times and weather date data to obtain the line load forecast results.

[0103] Figure 4 It is a structural diagram of a back propagation neural network provided according to an embodiment of the present invention. Figure 5 FIG. 1 is a flow chart of a load forecasting model obtained by training according to an embodiment of the present invention. Figure 4 and Figure 5 As shown in the figure, the process of training the load forecasting model is explained based on the structural characteristics of the back propagation neural network. Specifically:

[0104] Corresponding to the normalization preprocessing in the above step 2, the present invention takes into account that the units of the input data of each input node of the neural network may be different and the data range may differ greatly, and the weights of different input data of the neural network are proportional to the size of its own data, resulting in the effect of different variable data being too small or too large, which does not conform to the situation of actual load influencing factors. Therefore, the present invention uses the tansig() type transfer function as the transfer function of the hidden layer in the neural network, fixes the output result to (-1, 1), and the output result corresponding to the sample data needs to be adjusted to this range at the same time to facilitate comparison and learning of the neural network. Normalization preprocessing can be implemented based on the following formula:

[0105]

[0106] Among them, y represents the normalized output data, x max Represents the maximum data value in the sample data, x min represents the minimum data value in the sample data, and x represents the output data before normalization.

[0107] Furthermore, the load data of the original data range is recalculated and output through the output layer. The recalculation can be achieved based on the following formula:

[0108]

[0109] Among them, x max Represents the maximum data value in the sample data, x min represents the minimum data value in the sample data, and y represents the normalized preprocessing result.

[0110] Corresponding to the neural network training in step 3 above, assume that the number of input neurons is M, the number of hidden layer neurons is I, the number of output layer neurons is J, and the mth neuron in the input layer is denoted by x m , the i-th neuron in the hidden layer is denoted as k i , the jth neuron in the output layer is denoted as y i , connect x m and k i The weight is ω mi , connect k i and i The weight is ω ij The BPNN activation function tansig() is set to f(), and the output layer linear function purelin() is set to g().

[0111] Assume that the input of a single neuron is u and the output is v, such as Refers to the output of the first neuron in the hidden layer. The expected output of the network is:

[0112] d(n)=[d1,d2,...,d J ]

[0113] Among them, d(n) represents the expected output set of the network.

[0114] Specifically, the learning process of BPNN can be divided into two processes: forward propagation and error back propagation. The forward propagation process is as follows:

[0115] The output of the input layer is equal to the input signal of the entire network:

[0116] v m M (n) = x(n)

[0117] Among them, v m M(n) represents the output of the input layer, and x(n) represents the input signal of the entire network.

[0118] Furthermore, the input of the i-th neuron in the hidden layer is equal to v m The weighted sum of M(n):

[0119]

[0120] Among them, u i I(n) represents the input of the i-th neuron in the layer, ω mi Represents connection x m and k i The weight of . Furthermore, the output of the i-th neuron in the hidden layer is:

[0121] v i I (n) = f(u i I (n))

[0122] Among them, u i I(n) represents the input of the i-th neuron in the layer, v i I(n) represents the output of the i-th neuron in the hidden layer.

[0123] Furthermore, the input of the jth neuron in the output layer is equal to v i The weighted sum of I(n):

[0124]

[0125] Among them, u i J(n) represents the input of the jth neuron in the output layer, ω mi Represents connection x m and k i The weight, v iI(n) represents the output of the i-th neuron in the hidden layer.

[0126] Furthermore, the output of the jth neuron in the output layer is:

[0127] v j J (n) = g((u j J (n)))

[0128] Among them, v i J(n) represents the input of the jth neuron in the output layer, u i J(n) represents the input of the jth neuron in the output layer, and g() represents the output layer linear function purelin().

[0129] Furthermore, the error of the jth neuron in the output layer is:

[0130] e j (n) = d j (n)-v j J (n)

[0131] Among them, e j (n) represents the error of the jth neuron in the output layer, v i J(n) represents the input of the jth neuron in the output layer, d j (n) represents the expected output of the network for the jth neuron.

[0132] Furthermore, the total error of the network is:

[0133]

[0134] Among them, e(n) represents the total error of the network.

[0135] In general, the above steps are to pass the environmental input variables from the input layer → hidden layer → output layer layer by layer, and obtain the total error between the predicted output and the target output; use the total error to further improve the weight threshold matrix, calculate the influence of each weight and threshold on the total error from the last layer, and distribute the total error to different neurons according to the matrix parameters according to the influence size, and increase or decrease the parameter value through learning rules to reduce the error caused by itself. When the expected error or number of iterations is reached, the training is stopped, and the final training results are memorized in the weight threshold matrix of the network. When a new sample is input next time, the neural network will predict the output result through the final matrix.

[0136] Based on the technical solutions of the above embodiments, a high-precision load prediction model can be obtained, and the load prediction model can be used to perform high-precision line load prediction.

[0137] Embodiment 3

[0138] Figure 6 This is a schematic diagram of the structure of a line load prediction device provided by Embodiment 3 of the present invention. Figure 6 As shown, the device includes: a date and time determination module 310, a reference time determination module 320, a characteristic data determination module 330 and a line load prediction module 340.

[0139] Among them, the date and time determination module 310 is used to determine the real-time prediction date of the load prediction line and the prediction reference date corresponding to the real-time prediction date, and determine the real-time prediction time in the real-time prediction date; the reference time determination module 320 is used to determine multiple first reference times in the real-time prediction date and multiple second reference times in the real-time reference date according to the real-time prediction time; the characteristic data determination module 330 is used to determine the first load data corresponding to each of the first reference times and the second load data corresponding to each of the second reference times, and determine the weather date data corresponding to the real-time prediction date; the line load prediction module 340 is used to determine the load prediction result of the load prediction line at the real-time prediction time of the real-time prediction date according to the first load data corresponding to each of the first reference times, the second load data corresponding to each of the second reference times and the weather date data.

[0140] The technical solution of the embodiment of the present invention determines the real-time prediction date of the load prediction line and the prediction reference date corresponding to the real-time prediction date, determines the real-time prediction moment in the real-time prediction date; determines multiple first reference moments in the real-time prediction date and multiple second reference moments in the real-time reference date according to the real-time prediction moment; determines the first load data corresponding to each of the first reference moments and the second load data corresponding to each of the second reference moments, and determines the weather date data corresponding to the real-time prediction date; determines the load prediction result of the load prediction line at the real-time prediction moment of the real-time prediction date according to the first load data corresponding to each of the first reference moments, the second load data corresponding to each of the second reference moments, and the weather date data. The technical effect of improving the accuracy of line load prediction is achieved.

[0141] Optionally, the weather date data includes weather type, first temperature, second temperature and date type, and the multiple second reference moments in the real-time reference date correspond to the first reference moment and the real-time predicted moment in the real-time predicted date.

[0142] Optionally, the line load prediction module 340 is specifically used to:

[0143] The load forecasting model is used to perform load forecasting on the first load data corresponding to each first reference moment, the second load data corresponding to each second reference moment, the weather type, the first temperature, the second temperature and the date type, and determine the load forecast result of the load forecasting line at the real-time prediction moment on the real-time prediction date, wherein the load forecasting model is obtained by training a back propagation neural network based on sample data.

[0144] Optionally, the sample data includes training samples and test samples; the line load prediction device further includes: a neural network construction module, a neural network training module, a model testing module and a prediction model determination module;

[0145] The neural network construction module is used to determine the number of input nodes and the number of output nodes of the back propagation neural network before performing load forecasting on the first load data corresponding to each first reference moment, the second load data corresponding to each second reference moment, the weather type, the first temperature, the second temperature and the date type input by the load forecasting model, determine multiple candidate hidden layer node numbers according to the input node number and the output node number, and construct the back propagation neural network corresponding to each candidate hidden layer node number;

[0146] The neural network training module is used to set the number of input nodes, the number of output nodes and the transfer function of the hidden layer of the back-propagation neural network corresponding to each candidate hidden layer node number, and train the back-propagation neural network through the training samples to obtain a candidate prediction model;

[0147] The model testing module is used to test each candidate prediction model through the test sample to obtain a model testing result;

[0148] The prediction model determination module is used to determine the load prediction model among the candidate prediction models based on the model test results.

[0149] Optionally, the neural network training module includes: a neural network setting unit, used to set the neural transfer function as the transfer function of the hidden layer in the back-propagation neural network.

[0150] Optionally, the neural network construction module includes: a node number determination unit, used to determine the number of input nodes of the back propagation neural network based on preset reference time quantity information and preset feature data quantity information related to the weather date.

[0151] Optionally, the load forecast result includes line overload and / or line non-overload, and the line load forecasting device further includes: an overload warning module, which is used to issue a line overload warning through a target warning device after determining the load forecast result of the load forecast line at the real-time prediction time on the real-time prediction date, if the load forecast result is that the line is overloaded.

[0152] The line load prediction device provided in the embodiment of the present invention can execute the line load prediction method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0153] Embodiment 4

[0154] Figure 7 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0155] like Figure 7 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0156] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0157] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a line load prediction method.

[0158] In some embodiments, the line load prediction method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the line load prediction method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the line load prediction method in any other appropriate manner (e.g., by means of firmware).

[0159] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0160] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0161] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0162] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0163] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0164] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0165] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0166] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A line load prediction method, characterized in that: include: Determine a real-time prediction date of a load prediction line and a prediction reference date corresponding to the real-time prediction date, and determine a real-time prediction moment in the real-time prediction date; Determine a plurality of first reference moments in the real-time prediction date and a plurality of second reference moments in the real-time reference date according to the real-time prediction moment; Determine the first load data corresponding to each of the first reference moments and the second load data corresponding to each of the second reference moments, and determine the weather date data corresponding to the real-time forecast date; The load forecast result of the load forecast line at the real-time forecast moment on the real-time forecast date is determined based on the first load data corresponding to each of the first reference moments, the second load data corresponding to each of the second reference moments and the weather date data.

2. The method according to claim 1, characterized in that The weather date data includes a weather type, a first temperature, a second temperature and a date type, and the plurality of second reference moments in the real-time reference date correspond to the first reference moment and the real-time predicted moment in the real-time predicted date.

3. The method according to claim 2, characterized in that The step of determining the load forecast result of the load forecast line at the real-time forecast time on the real-time forecast date according to the first load data corresponding to each of the first reference times, the second load data corresponding to each of the second reference times, and the weather date data comprises: The load forecasting model is used to perform load forecasting on the first load data corresponding to each first reference moment, the second load data corresponding to each second reference moment, the weather type, the first temperature, the second temperature and the date type, and determine the load forecast result of the load forecasting line at the real-time prediction moment on the real-time prediction date, wherein the load forecasting model is obtained by training a back propagation neural network based on sample data.

4. The method according to claim 3, characterized in that The sample data includes training samples and test samples; before the load forecasting is performed on the first load data corresponding to each first reference time, the second load data corresponding to each second reference time, the weather type, the first temperature, the second temperature and the date type input by the load forecasting model, the method further includes: Determine the number of input nodes and the number of output nodes of the back propagation neural network, determine multiple candidate numbers of hidden layer nodes according to the number of input nodes and the number of output nodes, and construct the back propagation neural network corresponding to each candidate number of hidden layer nodes; For each of the back propagation neural networks corresponding to the candidate hidden layer node number, the number of input nodes, the number of output nodes and the transfer function of the hidden layer of the back propagation neural network are set, and the candidate prediction model is obtained by training the back propagation neural network through the training samples; For each of the candidate prediction models, testing the candidate prediction model using the test sample to obtain a model test result; The load forecasting model among the candidate forecasting models is determined based on the model testing result.

5. The method according to claim 4, characterized in that The step of setting the transfer function of the hidden layer of the back propagation neural network comprises: The neural transfer function is set as the transfer function of the hidden layer in the back-propagation neural network.

6. The method according to claim 4, characterized in that Determining the number of input nodes of the back propagation neural network includes: The number of input nodes of the back propagation neural network is determined according to the preset reference time quantity information and the preset characteristic data quantity information related to the weather date.

7. The method according to claim 1, characterized in that The load forecast result includes line overload and / or line non-overload, and after determining the load forecast result of the load forecast line at the real-time forecast time on the real-time forecast date, the method further includes: When the load forecast result indicates that the line is overloaded, a line overload warning is issued through a target warning device.

8. A line load prediction device, characterized in that: include: A date and time determination module, used to determine the real-time prediction date of the load prediction line and the prediction reference date corresponding to the real-time prediction date, and determine the real-time prediction time in the real-time prediction date; A reference time determination module, used to determine a plurality of first reference times in the real-time prediction date and a plurality of second reference times in the real-time reference date according to the real-time prediction time; a characteristic data determination module, used to determine the first load data corresponding to each of the first reference moments and the second load data corresponding to each of the second reference moments, and to determine the weather date data corresponding to the real-time forecast date; The line load prediction module is used to determine the load prediction result of the load prediction line at the real-time prediction moment of the real-time prediction date based on the first load data corresponding to each first reference moment, the second load data corresponding to each second reference moment and the weather date data.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the line load prediction method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the line load prediction method according to any one of claims 1 to 7 when executed.

Images