Intelligent analysis control system for power transmission operation

Abstract

The invention discloses an intelligent analysis control system for power transmission operation, and relates to the field of power transmission operation control. Historical power demand data and influence factor data are collected, corresponding timestamps are recorded respectively, and original data are obtained; preprocessing the original data to obtain first data, and dividing the first data into a training sample set and a verification sample set; a multi-layer perceptron model is constructed, and a training sample set is used for training the model; using the verification sample set to calculate a mean square error as an evaluation index to evaluate the accuracy of the model, and obtaining an optimal prediction model; inputting the obtained influence factor data corresponding to the prediction time point into the optimal prediction model, and calculating a power demand prediction value of the prediction time point; the maximum bearing capacity of the power transmission system is obtained, the load rate of the power transmission line is calculated according to the power demand prediction value at the prediction time point, and evaluation and early warning are conducted on the power transmission system.

Description

Technical Field

[0001] The present invention relates to the field of power transmission operation control, and in particular to a power transmission operation intelligent analysis and control system. Background Art

[0002] With the advancement of the construction of the power Internet of Things, the types of smart terminals deployed on transmission lines are gradually increasing and the coverage is constantly expanding. Traditional transmission system management methods are facing many challenges.

[0003] These challenges include how to accurately predict future electricity demand, how to monitor and evaluate the load of the transmission system in real time, and how to provide early warning of potential imbalances in electricity supply and demand. Existing solutions usually rely on manual experience and simple statistical methods, which are difficult to cope with complex and changing electricity markets and environmental conditions.

[0004] Therefore, it is particularly important to develop an intelligent analysis and control system that can automatically analyze historical data, predict future power demand, and monitor the status of the transmission system in real time. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent analysis and control system for power transmission operation to solve the deficiencies in the background technology.

[0006] The technical solution of the present invention is: an intelligent analysis and control system for power transmission operation, comprising: a data collection and preprocessing module, collecting historical power demand data and influencing factor data, recording corresponding timestamps respectively, and obtaining original data; preprocessing the original data to obtain first data, and dividing the first data into a training sample set and a verification sample set;

[0007] The power forecasting module builds a multi-layer perceptron model and uses the training sample set to train the model; the mean square error is calculated using the validation sample set as an evaluation indicator to evaluate the accuracy of the model and obtain the optimal forecasting model; the influencing factor data corresponding to the forecast time point is input into the optimal forecasting model to calculate the power demand forecast value at the forecast time point;

[0008] The evaluation and early warning module obtains the maximum carrying capacity of the transmission system, and evaluates and warns the transmission system by calculating the load rate of the transmission line based on the predicted value of power demand at the predicted time point.

[0009] Preferably, the raw data is preprocessed, including removing duplicate data, filling missing values ​​with the mean or median, unifying and aligning the timestamp formats of all raw data records, and normalizing the sorted data to obtain the first data.

[0010] Preferably, the normalization method is as follows: counting the maximum and minimum values ​​corresponding to different types of data in the original data, and performing normalization on different types of data respectively;

[0011] The calculation formula is:

[0012]

[0013] In the formula, x max and x min are the maximum and minimum values ​​corresponding to the same type of data, x' is the original data of the same type, and x is the normalized data.

[0014] Preferably, the implementation content of the power prediction module includes: constructing a multi-layer perceptron model and training the model using a training sample set; wherein the multi-layer perceptron model includes an input layer, a hidden layer, and an output layer;

[0015] Assume that the input feature vector is X = [y, T, H, D]; the output layer has 4 neuron nodes, each corresponding to an input feature; where y is the historical power demand data sequence, T is the weather temperature, H is the weather humidity, and D is the date type;

[0016] For the calculation from the input layer to the first hidden layer;

[0017]

[0018] In the formula, is the output vector of the first hidden layer; f() represents the activation function, and the ReLU function is commonly used; W 1 represents the weight matrix of the first hidden layer; b 1 represents the bias vector of the first hidden layer; X i Represents the feature vector corresponding to the i-th time point;

[0019] Calculation from the first hidden layer to the second hidden layer:

[0020]

[0021] In the formula, is the output vector of the second hidden layer; f() represents the activation function, and the ReLU function is commonly used; W 2 represents the weight matrix of the second hidden layer; b 2 represents the bias vector of the second hidden layer;

[0022] Calculation from the second hidden layer to the output layer:

[0023]

[0024] In the formula, represents the predicted value of power demand at the i-th time point; W 3 represents the weight matrix of the output layer; b 3 Represents the bias vector of the output layer;

[0025] For each sample in the training sample set, the input features of the sample are input into the model, and the power demand forecast value is calculated according to the above model formula.

[0026] Preferably, the calculation formula for defining the loss function L is as follows:

[0027]

[0028] Where RSS represents the loss function; y i represents the power demand value at the i-th time point; represents the predicted value of electricity demand at the i-th time point;

[0029] By calculating the partial derivative RSS′ of the RSS loss function with respect to the weight and bias term of the model and setting the partial derivative RSS′=0, the RSS loss function is minimized, the weight and bias term of the model are obtained, and the multi-layer perceptron model is determined.

[0030] Preferably, the data of the verification sample set is substituted into the trained multi-layer perceptron model to obtain the power demand forecast value, and the accuracy of the model is evaluated by calculating the mean square error as an evaluation index to obtain the optimal prediction model;

[0031] The formula for calculating the mean square error (MSE) is:

[0032]

[0033] In the formula, MSE represents the mean square error; y m It represents the actual value of power demand in the verification sample set; represents the predicted value of power demand in the validation sample set; m = 1, 2, ..., M, where M represents the total number of validation sample sets;

[0034] Set the model error threshold and compare the calculated mean square error value MSE with the model error threshold. If MSE ≤ model error threshold, it means that the accuracy of the model meets the standard and the corresponding model is obtained as the optimal prediction model. If MSE > model error threshold, it means that the prediction ability of the model is poor and the model parameters need to be optimized.

[0035] Preferably, the current time point t and the prediction step length r are obtained, the influencing factor data corresponding to the prediction time point t+r are obtained, input into the optimal prediction model, and the power demand prediction value at the prediction time point t+r is obtained.

[0036] Preferably, the implementation content of the evaluation and early warning module includes: obtaining the maximum carrying capacity of the power transmission system, and calculating the load rate of the power transmission line according to the power demand forecast value at the forecast time point;

[0037] The calculation formula is as follows:

[0038]

[0039] Where LR is the load factor of the transmission line; is the predicted value of power demand at the prediction time point t+r; C is the maximum carrying capacity;

[0040] An assessment and warning is made based on the calculated load rate. If LR≤80%, it indicates that the transmission system is in a safe state and no additional measures are needed. If 80%<LR<100%, it indicates that the transmission system is close to full load operation, an orange warning signal is issued and the emergency dispatch plan is urgently initiated. If LR≥100%, it indicates that the load of the transmission system exceeds its carrying capacity, a red warning signal is issued and emergency measures are taken immediately.

[0041] The present invention also provides a method for intelligent analysis and control of power transmission operation, comprising the following steps:

[0042] Collect historical power demand data and influencing factor data, record corresponding timestamps respectively, and obtain original data; pre-process the original data to obtain first data, and divide the first data into a training sample set and a verification sample set;

[0043] Construct a multi-layer perceptron model and use the training sample set to train the model; use the validation sample set to calculate the mean square error as an evaluation indicator to evaluate the accuracy of the model and obtain the optimal prediction model;

[0044] The acquired influencing factor data corresponding to the forecast time point is input into the optimal forecast model to calculate the power demand forecast value at the forecast time point;

[0045] Obtain the maximum carrying capacity of the power system, evaluate and warn the transmission system based on the predicted power demand at the predicted time point and by calculating the load rate of the transmission line.

[0046] Compared with the existing solutions, the present invention achieves the following beneficial effects:

[0047] The present invention collects historical power demand data and influencing factor data, respectively records corresponding timestamps, and obtains original data; pre-processes the original data to obtain first data, and divides the first data into a training sample set and a verification sample set;

[0048] The multi-layer perceptron model constructed by the present invention through machine learning algorithm can more accurately predict future power demand, with higher accuracy than traditional methods. The entire process of model training and prediction is completed automatically, reducing the need for manual intervention and improving work efficiency.

[0049] The system of the present invention can monitor the load condition of the power transmission system in real time and issue early warning according to the prediction results, which helps to avoid system failure or power outage caused by imbalance in power supply and demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The present invention will be further described below in conjunction with the accompanying drawings.

[0051] Figure 1 This is a module block diagram of an intelligent analysis and control system for power transmission operation according to the present invention;

[0052] Figure 2 The present invention is a flowchart of a method for intelligent analysis and control of power transmission operation. DETAILED DESCRIPTION

[0053] The technical solutions in the embodiments of the present invention will be described clearly and completely 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, rather than all the embodiments.

[0054] Embodiment 1, as Figure 1 As shown, a power transmission operation intelligent analysis and control system includes: a data acquisition and preprocessing module, a power prediction module and an evaluation and early warning module;

[0055] The data collection and preprocessing module collects historical power demand data and influencing factor data, records the corresponding timestamps, and obtains the original data; the influencing factor data includes but is not limited to weather temperature, weather humidity, and date type;

[0056] Preprocessing the raw data includes removing duplicate data, filling missing values ​​with the mean value or the median, unifying and aligning the timestamp formats of all raw data records, and normalizing the sorted data to obtain the first data;

[0057] Furthermore, the maximum and minimum values ​​corresponding to different types of data in the original data are counted, and the different types of data are normalized respectively;

[0058] The calculation formula is:

[0059]

[0060] In the formula, x max and x minare the maximum and minimum values ​​corresponding to the same type of data, x' is the original data of the same type, and x is the normalized data;

[0061] Dividing the first data into a training sample set and a validation sample set according to a certain ratio; wherein the ratio is usually set to 8:2;

[0062] In the implementation manner of the present invention, the original data is preprocessed and different types of data are normalized, which is beneficial to improving the convergence speed and prediction accuracy in the model training process and reducing the deviation in model training; the first data is divided into a training sample set and a verification sample set, wherein the training sample set is used to train the model, and the verification sample set is used to evaluate the performance and generalization ability of the model.

[0063] The power forecasting module traverses the sample training set to obtain historical power demand data and influencing factor data; let the historical power demand data sequence be y = {yi}, where y i represents the power demand value at the i-th time point, i = 1, 2, ..., n, and n is the total number of data points;

[0064] In the corresponding influencing factor data, the weather temperature is recorded as T = {T i}, represents the weather temperature at the i-th time point; the weather humidity is recorded as H = {H i}, indicating the weather humidity at the i-th time point; the date type is recorded as D = {D i}, indicating the date type of the i-th time point; where D i =1 means working day, D i =0 means holidays;

[0065] Build a multi-layer perceptron model and train the model using a training sample set;

[0066] Furthermore, the multi-layer perceptron model includes an input layer, a hidden layer, and an output layer;

[0067] Assume the input feature vector is X = [y, T, H, D]; the output layer has 4 neuron nodes, each corresponding to an input feature;

[0068] For the calculation from the input layer to the first hidden layer;

[0069]

[0070] In the formula, is the output vector of the first hidden layer; f() represents the activation function, and the ReLU function is commonly used; W 1 represents the weight matrix of the first hidden layer; b 1 represents the bias vector of the first hidden layer; X iRepresents the feature vector corresponding to the i-th time point;

[0071] Calculation from the first hidden layer to the second hidden layer:

[0072]

[0073] In the formula, is the output vector of the second hidden layer; f() represents the activation function, and the ReLU function is commonly used; W 2 represents the weight matrix of the second hidden layer; b 2 represents the bias vector of the second hidden layer;

[0074] Calculation from the second hidden layer to the output layer:

[0075]

[0076] In the formula, represents the predicted value of power demand at the i-th time point; W 3 represents the weight matrix of the output layer; b 3 Represents the bias vector of the output layer;

[0077] For each sample in the training sample set, the input features of the sample are input into the model, and the power demand forecast value is calculated according to the above model formula;

[0078] Define the loss function L to measure the difference between the model prediction value and the true value;

[0079] The calculation formula is as follows:

[0080]

[0081] Where RSS represents the loss function; y i represents the power demand value at the i-th time point; represents the predicted value of electricity demand at the i-th time point;

[0082] By finding the partial derivative RSS′ of the RSS loss function with respect to the weight and bias term of the model, and setting the partial derivative RSS′=0, the RSS loss function is minimized, and the weight and bias term of the model are obtained, thereby determining the multi-layer perceptron model;

[0083] Furthermore, the data of the validation sample set is substituted into the trained multi-layer perceptron model to obtain the power demand forecast value. The accuracy of the model is evaluated by calculating the mean square error as an evaluation index to obtain the optimal prediction model.

[0084] The formula for calculating the mean square error (MSE) is:

[0085]

[0086] In the formula, MSE represents the mean square error; y m It represents the actual value of power demand in the verification sample set; represents the predicted value of power demand in the validation sample set; m = 1, 2, ..., M, where M represents the total number of validation sample sets;

[0087] It should be noted that the smaller the MSE value, the smaller the deviation between the model's prediction results and the actual data, and the higher the accuracy of the model; if the MSE value is large, it means that the model's prediction ability is poor, and the model needs to be improved;

[0088] Set the model error threshold, compare the calculated mean square error value MSE with the model error threshold. If MSE ≤ model error threshold, it means that the accuracy of the model meets the standard, and the corresponding model is obtained as the optimal prediction model. If MSE > model error threshold, it means that the prediction ability of the model is poor, and the model parameters need to be optimized. The setting of the model error threshold depends on the specific application scenario and business needs.

[0089] Further, the current time point t and the prediction step length r are obtained, the influencing factor data corresponding to the prediction time point t+r are obtained, and input into the optimal prediction model to obtain the power demand prediction value at the prediction time point t+r;

[0090] In the implementation steps of the present invention, the power demand value at a certain point in the future is predicted by constructing an optimal prediction model, which is beneficial to the reasonable allocation of power resources and improves the operating efficiency of the power transmission system.

[0091] The evaluation and early warning module obtains the maximum carrying capacity of the transmission system and calculates the load rate of the transmission line based on the power demand forecast value at the forecast time point; the maximum carrying capacity refers to the maximum power load that the power system can safely and stably bear;

[0092] The calculation formula is as follows:

[0093]

[0094] Where LR is the load factor of the transmission line; is the predicted value of power demand at the prediction time point t+r; C is the maximum carrying capacity;

[0095] Evaluation and warning are carried out based on the calculated load rate. If LR≤80%, the transmission system is in a safe state and no additional measures are required. If 80%<LR<100%, it indicates that the transmission system is close to full load operation, an orange warning signal is issued and the emergency dispatch plan is urgently initiated. If LR≥100%, it indicates that the load of the transmission system exceeds its carrying capacity, a red warning signal is issued and emergency measures are taken immediately.

[0096] The emergency dispatch plan includes, but is not limited to, giving priority to ensuring the power supply to key users (such as hospitals, transportation hubs, etc.); arranging to increase power generation capacity as soon as possible, such as starting fast-response power generation equipment or urgently deploying power from other areas;

[0097] Emergency measures include but are not limited to cutting off some non-critical loads and implementing an orderly power consumption plan. At the same time, we will make every effort to increase power generation capacity and deploy power resources;

[0098] In the implementation steps of the present invention, based on the power demand forecast, it is evaluated whether the transmission capacity of the existing power grid can meet future demand, and the transportation status of the transmission system is monitored in real time. By intelligently analyzing the load rate, abnormal situations are discovered and early warning signals are issued, and corresponding measures are taken to reduce the risk of accidents, thereby ensuring the safety and stability of power grid operation.

[0099] Embodiment 2, as Figure 2 As shown, a method for intelligent analysis and control of power transmission operation comprises the following steps:

[0100] Collect historical power demand data and influencing factor data, record corresponding timestamps respectively, and obtain original data; pre-process the original data to obtain first data, and divide the first data into a training sample set and a verification sample set;

[0101] Construct a multi-layer perceptron model and use the training sample set to train the model; use the validation sample set to calculate the mean square error as an evaluation indicator to evaluate the accuracy of the model and obtain the optimal prediction model;

[0102] The acquired influencing factor data corresponding to the forecast time point is input into the optimal forecast model to calculate the power demand forecast value at the forecast time point;

[0103] Obtain the maximum carrying capacity of the transmission system, evaluate and warn the transmission system by calculating the load rate of the transmission line based on the predicted value of power demand at the predicted time point.

[0104] In the several embodiments provided by the present invention, it should be understood that the disclosed system can be implemented in other ways. For example, the above-described embodiments of the invention are only illustrative, for example, the division of modules is only a logical function division, and there may be other division methods in actual implementation.

[0105] The modules described as separate accessories may or may not be physically separated, and the accessories shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0106] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0107] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. An intelligent analysis and control system for power transmission operation, characterized in that: include: The data collection and preprocessing module collects historical power demand data and influencing factor data, records the corresponding timestamps, and obtains the original data; Preprocessing the original data to obtain first data, and dividing the first data into a training sample set and a verification sample set; The power forecasting module builds a multi-layer perceptron model and uses the training sample set to train the model; the mean square error is calculated using the validation sample set as an evaluation indicator to evaluate the accuracy of the model and obtain the optimal forecasting model; the influencing factor data corresponding to the forecast time point is input into the optimal forecasting model to calculate the power demand forecast value at the forecast time point; The evaluation and early warning module obtains the maximum carrying capacity of the transmission system, and evaluates and warns the transmission system by calculating the load rate of the transmission line based on the predicted value of power demand at the predicted time point.

2. The intelligent analysis and control system for power transmission operation according to claim 1, characterized in that: The raw data is preprocessed, including removing duplicate data, filling missing values ​​with the mean or median, unifying and aligning the timestamp formats of all raw data records, and normalizing the sorted data to obtain the first data.

3. The intelligent analysis and control system for power transmission operation according to claim 2, characterized in that: The normalization method is as follows: count the maximum and minimum values ​​corresponding to different types of data in the original data, and perform normalization on different types of data respectively; The calculation formula is: In the formula, x max and x min are the maximum and minimum values ​​corresponding to the same type of data, x′ is the original data of the same type, and x is the normalized data.

4. The intelligent analysis and control system for power transmission operation according to claim 1, characterized in that: The implementation content of the power forecasting module includes: constructing a multi-layer perceptron model and training the model using a training sample set; wherein the multi-layer perceptron model includes an input layer, a hidden layer, and an output layer; Assume that the input feature vector is X = [y, T, H, D]; the output layer has 4 neuron nodes, each corresponding to an input feature; where y is the historical power demand data sequence, T is the weather temperature, H is the weather humidity, and D is the date type; For the calculation from the input layer to the first hidden layer; In the formula, is the output vector of the first hidden layer; f() represents the activation function; W 1 represents the weight matrix of the first hidden layer; b 1 represents the bias vector of the first hidden layer; X i Represents the feature vector corresponding to the i-th time point; Calculation from the first hidden layer to the second hidden layer: In the formula, is the output vector of the second hidden layer; W 2 represents the weight matrix of the second hidden layer; b 2 represents the bias vector of the second hidden layer; Calculation from the second hidden layer to the output layer: In the formula, represents the predicted value of power demand at the i-th time point; W 3 represents the weight matrix of the output layer; b 3 Represents the bias vector of the output layer; For each sample in the training sample set, the input features of the sample are input into the model, and the power demand forecast value is calculated according to the above model formula.

5. The intelligent analysis and control system for power transmission operation according to claim 4, characterized in that: The calculation formula for defining the loss function L is as follows: Where RSS represents the loss function; y i represents the power demand value at the i-th time point; represents the predicted value of electricity demand at the i-th time point; By calculating the partial derivative RSS′ of the RSS loss function with respect to the weight and bias term of the model and setting the partial derivative RSS′=0, the RSS loss function is minimized, the weight and bias term of the model are obtained, and the multi-layer perceptron model is determined.

6. The intelligent analysis and control system for power transmission operation according to claim 5, characterized in that: Substitute the data of the validation sample set into the trained multi-layer perceptron model to obtain the power demand forecast value. The accuracy of the model is evaluated by calculating the mean square error as an evaluation indicator to obtain the optimal prediction model. The formula for calculating the mean square error is: In the formula, MSE represents the mean square error; y m It represents the actual value of power demand in the verification sample set; represents the predicted value of power demand in the validation sample set; m = 1, 2, ..., M, where M represents the total number of validation sample sets; Set the model error threshold and compare the calculated mean square error value MSE with the model error threshold. If MSE ≤ model error threshold, it means that the accuracy of the model meets the standard and the corresponding model is obtained as the optimal prediction model. If MSE > model error threshold, it means that the prediction ability of the model is poor and the model parameters need to be optimized.

7. The intelligent analysis and control system for power transmission operation according to claim 6, characterized in that: Get the current time point t and the prediction step length r, get the influencing factor data corresponding to the prediction time point t+r, input them into the optimal prediction model, and get the power demand prediction value at the prediction time point t+r.

8. The intelligent analysis and control system for power transmission operation according to claim 7, characterized in that: The implementation of the evaluation and early warning module includes: obtaining the maximum carrying capacity of the transmission system, calculating the load rate of the transmission line based on the predicted value of power demand at the predicted time point; The calculation formula is as follows: Where LR is the load factor of the transmission line; is the predicted value of power demand at the prediction time point t+r; C is the maximum carrying capacity; Evaluation and warning are carried out based on the calculated load rate. If LR≤80%, it indicates that the transmission system is in a safe state and no additional measures are needed. If 80%<LR<100%, it indicates that the transmission system is close to full load operation, an orange warning signal is issued and the emergency dispatch plan is urgently initiated. If LR≥100%, it indicates that the load of the transmission system exceeds its carrying capacity, a red warning signal is issued and emergency measures are taken immediately.

Images