Power grid load optimization intelligent control system and method

Abstract

The invention relates to the technical field of power grid control systems, in particular to a power grid load optimization intelligent control system and method, and the system comprises a power grid load data collection module which collects power grid load data; the power grid load curve analysis and prediction module is used for making a power grid load curve according to the power grid load data and predicting the power grid load curve to obtain a future power grid load curve; the dispatching strategy system module is used for carrying out regional division on an industrial electricity utilization region according to a future power grid load curve to obtain divided regions, and carrying out power distribution according to the divided regions; and the real-time monitoring and adjusting module is used for monitoring the real-time operation state of the equipment in each divided area and optimizing power distribution according to the operation state of the equipment in each divided area. According to the invention, by collecting and monitoring the power grid load data, a reasonable power distribution scheme is provided, and power resources are reasonably distributed.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid control systems, and particularly to an intelligent control system and method for optimizing power grid loads. Background Art

[0002] Smart grid technology refers to the use of advanced information and communication technologies and sensor technologies to achieve intelligent management and optimal allocation of power systems. The development of this technology provides strong technical support for power grid load optimization systems. Smart grid technology can collect, analyze, and control power load data in real time, improving the response speed and efficiency of power systems; at the same time, smart grid technology can also achieve intelligent management and optimal scheduling of power loads, improving the power supply quality and efficiency of power systems.

[0003] Chinese Patent Publication No.: CN109950905B discloses a power grid load statistics method. This method can process a large amount of power grid operation data, quickly mine, clean, and analyze massive data through big data technology, greatly improving the efficiency of load statistics. Compared with traditional manual statistics or statistics methods based on small sample data, it can provide decision-making support for power grid operation and management more timely. It has good adaptability to power grids of different regions, scales, and structures. Whether it is an urban power grid, a rural power grid, or a large regional power grid, as long as there is sufficient data support, this method can perform customized load statistics and analysis according to the actual operation conditions of the power grid, providing a powerful tool for the refined management of the power grid. However, this method largely relies on historical data to establish statistical models and predict future loads, lacks real-time data feedback, and does not elaborate on how to seamlessly connect and cooperate the power system and equipment after monitoring the electrical load data, and still cannot meet the development needs of future power systems. Summary of the Invention

[0004] Therefore, the present invention provides an intelligent control system and method for optimizing power grid loads to overcome the problems of power waste and overload in the power grid system due to unreasonable power resource allocation in the prior art.

[0005] To achieve the above object, the present invention provides an intelligent control system and method for optimizing power grid loads, including: A power grid load data acquisition module for acquiring power grid load data in power grid equipment; A power grid load curve analysis and prediction module for making a power grid load curve according to the power grid load data, performing anomaly judgment according to the power grid load curve, and also for predicting the power grid load curve to obtain a future power grid load curve; The scheduling strategy system module is used to divide the industrial electricity consumption area according to the future power grid load curve, obtain the divided areas, and allocate power according to the divided areas; The real-time monitoring and adjustment module is used to monitor the real-time operating status of the devices in each divided area, optimize the power allocation according to the operating status of the devices in each divided area, and also correct the optimization process of power allocation according to the device power factor.

[0006] Furthermore, the power grid load curve analysis and prediction module calculates the power load P = {p1, p2, p3... pn} corresponding to each time point according to the voltage U = {u1, u2, u3... un} corresponding to each time point, the current A = {a1, a2, a3... an} corresponding to each time point, and the load power factor cosβ in the power grid load data, sets P = UAcosβ, and makes a power grid load curve according to the power load P corresponding to the time point.

[0007] Furthermore, the power grid load curve analysis and prediction module is used to compare the power load Pi corresponding to each time point in the power grid load curve with the preset power load P0, judge the power load situation at this time point according to the comparison result, and predict the power grid load curve according to the judgment result, where: When P1 ≤ P0, the power grid load curve analysis and prediction module determines that the power load situation at this time point is in a normal state; When P1 > P0, the power grid load curve analysis and prediction module determines that the power load situation at this time point is in an abnormal state and predicts the power grid load curve.

[0008] Further, when the power grid load curve analysis and prediction module predicts the power grid load curve, the power grid load curve analysis and prediction module selects a convolutional neural network model as the basic framework of the future load curve prediction model, divides 70% of the historical power grid load curve data set into a prediction training set, 15% of the historical power grid load curve data set into a prediction validation set, and 15% of the historical power grid load curve data set into a prediction test set. The convolutional neural network model is trained according to the prediction training set, the weights of the convolutional neural network model are updated through the backpropagation algorithm, and after each epoch, the prediction validation set is used to validate the convolutional neural network model to obtain the prediction validation loss value and the prediction validation accuracy. When the validation loss value and the validation accuracy meet the preset prediction conditions, the convolutional neural network model is tested according to the prediction test set to obtain the prediction test accuracy. When the prediction test accuracy reaches the preset accuracy, the convolutional neural network model is output as the future load curve prediction model. The power grid load curve analysis and prediction module converts the power grid load curve into a picture form and inputs it into the future load curve prediction model, and obtains the future power grid load curve output by the future load curve prediction model.

[0009] Further, the dispatching strategy system module compares the total power consumption load PR of each region in the future power grid load curve with the preset maximum power grid load PR0max and the preset minimum power grid load PR0min, judges the regional importance level of this region according to the comparison result, and divides the region according to the judgment result, where: When PR≥PR0max, the regional division unit determines that the regional importance level of this region is important and divides this region into an important load area; When PR0min<PR<PR0max, the regional division unit determines that the regional importance level of this region is secondary and divides this region into an adjustable load area; When PR≤PR0min, the regional division unit determines that the regional importance level of this region is unimportant and divides this region into a general load area.

[0010] Further, the dispatching strategy system module distributes power according to the divided regions, where: For the important load area, according to the number of devices m in the important load area in the power grid load data and the rated power pj of the devices in the important load area, calculate the basic power distribution amount Pb of the important load area, and set , and calculate the actual power distribution amount Ps of the important load area according to Pb and the redundancy coefficient ω, and set Ps = ωPb, 1.1<ω<1.3; Adjustable load area: Calculate the power distribution amount Pq of the adjustable load area according to the rated power pk of the equipment in the adjustable load area in the grid load data, the grid load adjustment coefficient μ, and the equipment adjustment flexibility coefficient Fk, and set Pq = pk×(1 - μ×Fk); General load area: Calculate the power supply amount Pf of the general load area according to the actual power distribution amount Ps of the important load area, the power distribution amount Pq of the adjustable load area, and the total grid power supply amount Pt, and set Pf = Pt - (Pq + Ps).

[0011] Furthermore, the real-time monitoring and adjustment module compares the total real-time power Q of all areas in the grid load data with the preset power Q0, judges the power supply and demand situation of all areas according to the comparison result, and optimizes the power supply and demand situation of all areas according to the judgment result, where: When Q ≥ Q0, the real-time monitoring and adjustment module determines that the power supply and demand situation of all areas is sufficient, and does not optimize the power supply and demand situation of all areas; When Q < Q0, the real-time monitoring and adjustment module determines that the power supply and demand situation of all areas is insufficient, and optimizes the power supply and demand situation of all areas. The optimization method is to recalculate the actual power distribution amount of the important load area, the power distribution amount of the adjustable load area, and the power supply amount of the general load area according to the total grid power supply amount Pt, the first feedback coefficient f, the second feedback coefficient b, and the third feedback coefficient c, and obtain the adjusted actual power distribution amount Ps' of the important load area, the adjusted power distribution amount Pq' of the adjustable load area, and the adjusted power supply amount Pf' of the general load area. Set Ps' = f×Pt, Pq' = b×Pt, Ps' = c×Pt, f = 0.7, b = 0.3, c = 0.1.

[0012] Furthermore, the real-time monitoring and adjustment module collects the voltages of each area under the adjusted actual power distribution amount of the important load area, the adjusted power distribution amount of the adjustable load area, and the adjusted power supply amount of the general load area in real time, compares the collected total voltage UZ with the preset total voltage UZ0, and judges the equipment failure state according to the comparison result, where: When UZ ≤ UZ0, the real-time monitoring and adjustment module determines that the equipment failure state is normal; When UZ > UZ0, the real-time monitoring and adjustment module determines that the equipment failure state is abnormal.

[0013] Furthermore, the real-time monitoring and adjustment module also calculates the equipment power factor cosλ according to the active power S and reactive power D of the equipment in the grid load data, and sets , compare cosλ with the preset power factor cosλ0, judge the reactive power compensation situation of the device according to the comparison result, and correct the judgment situation of the power supply and demand situation in this area according to the judgment result. Among them, When cosλ≥cosλ0, the real-time monitoring and adjustment module determines that the reactive power situation of the device is sufficient and does not correct the judgment situation of the power supply and demand situation in this area; When cosλ<cosλ0, the real-time monitoring and adjustment module determines that the reactive power compensation situation of the device is insufficient and corrects the judgment situation of the power supply and demand situation in this area through the correction coefficient =0.8 + 0.2e -（cosλ0-cosλ） , cosλ>0, correct the preset power to obtain the corrected preset power Q0b, and set Q0b = ×Q0, and compare the corrected preset power Q0b with the real-time power Q of each area again, and re-judge the power supply and demand situation in this area according to the comparison result.

[0014] The present invention also provides an intelligent control method for optimizing the grid load. The method includes: Step S1: Collect the grid load data in the grid equipment; Step S2: Make a grid load curve according to the collected grid load data, judge the abnormality of the grid load curve, and use the future load curve prediction model to predict the future trend of the abnormal grid load curve to obtain the future grid load curve; Step S3: Divide the industrial electricity consumption area according to the total electricity consumption load of each area in the future grid load curve to obtain the divided area; Step S4: Monitor the real-time operating status of the devices in different areas and optimize the power distribution according to the real-time operating status; Step S5: Calculate the power factor of the device, judge the reactive power compensation situation of the device according to the power factor of the device, and correct the optimization process of the power distribution according to the judgment result.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: by real-time monitoring the energy consumption of the device, intelligent scheduling is carried out according to different production requirements, power resources are reasonably allocated, power waste and overload phenomena are avoided, the continuity and stability of the production process are ensured, and through data analysis and prediction, optimization suggestions are provided to help enterprises reduce energy costs and improve production efficiency. Among them, the system collects the grid load data in the grid equipment through the grid load data acquisition module for subsequent analysis based on the grid load data. The system makes a grid load curve according to the grid load data through the grid load curve analysis and prediction module, and makes an abnormality judgment according to the made grid load curve. When the grid load curve is abnormal, the future grid load curve is predicted. Predicting the grid load curve can understand the change trend of industrial electricity load in advance and provide a reference for grid scheduling and enterprise production plan adjustment. The system divides the industrial electricity consumption area through the scheduling strategy system module for power distribution in the divided area. Power distribution according to the area can improve the utilization efficiency of grid equipment. The system monitors the divided area through the real-time monitoring and adjustment module, monitors its real-time operation status, and further optimizes the power distribution according to its operation status. By calculating the power factor of the equipment, the judgment of the regional power supply and demand situation is corrected, and the power distribution is further refined to ensure the rationality of the power distribution. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a schematic structural diagram of the grid load optimization intelligent control system of this embodiment; Figure 2 is a schematic flow diagram of the grid load optimization intelligent control method of this embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] In order to make the objectives and advantages of the present invention clearer, the present invention will be further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0018] The preferred embodiments of the present invention will be described below with reference to the drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principle of the present invention and do not limit the protection scope of the present invention.

[0019] It should be noted that in the description of the present invention, the terms indicating the direction or positional relationship such as "upper", "lower", "left", "right", "inner", "outer", etc. are based on the direction or positional relationship shown in the drawings. This is only for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention.

[0020] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and defined, the terms "installation", "connection", and "coupling" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the communication inside two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0021] Please refer to Figure 1 as shown, which is a schematic structural diagram of the intelligent control system for optimizing the power grid load in this embodiment. The system includes: A power grid load data acquisition module for acquiring the power grid load data in power grid equipment; A power grid load curve analysis and prediction module for making a power grid load curve based on the power grid load data, performing anomaly judgment according to the power grid load curve, and also predicting the power grid load curve to obtain a future power grid load curve; A dispatching strategy formulation module for dividing the industrial power consumption area according to the future power grid load curve to obtain the divided areas, and performing power distribution according to the divided areas; A real-time monitoring and adjustment module for monitoring the real-time operating status of the equipment in each divided area, optimizing the power distribution according to the operating status of the equipment in each divided area, and also correcting the optimization process of the power distribution according to the equipment power factor. Specifically, the system is applied to the power management terminal in industrial production. By real-time monitoring the energy consumption of equipment, it conducts intelligent scheduling according to different production demands, reasonably allocates power resources, avoids power waste and overload phenomena, and ensures the continuity and stability of the production process. Through data analysis and prediction, it provides optimization suggestions to help enterprises reduce energy costs and improve production efficiency. Among them, the system collects the grid load data in grid equipment through the grid load data collection module for subsequent analysis based on the grid load data. The system makes a grid load curve according to the grid load data through the grid load curve analysis and prediction module, and conducts anomaly judgment based on the made grid load curve. When the grid load curve is abnormal, it predicts the future grid load curve. Predicting the grid load curve can understand the change trend of industrial electricity load in advance and provide reference for grid scheduling and enterprise production plan adjustment. The system divides the industrial electricity consumption area through the scheduling strategy system module for power distribution in the divided areas. Power distribution according to the area can improve the utilization efficiency of grid equipment. The system monitors the divided areas through the real-time monitoring and adjustment module, monitors their real-time operating status, and further optimizes the power distribution according to their operating status. By calculating the equipment power factor, it corrects the judgment of the regional power supply and demand situation and further refines the power distribution to ensure the rationality of power distribution.

[0022] Specifically, the grid load data collected by the grid load data collection module includes the voltage corresponding to each time point, the current corresponding to each time point, the number of devices in the important load area, the rated power of the devices in the important load area, the rated power of the devices in the adjustable load area, the total grid power supply, the real-time total power of all areas, the adjusted voltage of the important load area, the adjusted voltage of the adjustable load area, the adjusted fault voltage of the general load area, the active power of the device, and the reactive power of the device. The grid load data collection module collects the voltage corresponding to each time point through a voltage meter. The grid load data collection module collects the current corresponding to each time point through an ammeter. The grid load data collection module collects the number of devices in the important load area, the rated power of the devices in the important load area, and the rated power of the devices in the adjustable load area through the equipment registration records in the grid coverage area. The grid load data collection module collects the total grid power supply through a watt-hour meter. The grid load data collection module collects the current and voltage through a voltmeter and an ammeter, and calculates the real-time total power of all areas according to the current and voltage. The grid load data collection module collects the adjusted voltage of the important load area, the adjusted voltage of the adjustable load area, and the adjusted fault voltage of the general load area through a voltmeter. The grid load data collection module collects the current and voltage through an ammeter and a voltmeter, and calculates the active power and reactive power of the device according to the current and voltage.

[0023] Specifically, the power grid load curve analysis and prediction module calculates the power load P = {p1, p2, p3... pn} corresponding to each time point according to the voltage U = {u1, u2, u3... un} corresponding to each time point, the current A = {a1, a2, a3... an} corresponding to each time point, and the load power factor cosβ in the power grid load data, sets P = UAcosβ, and makes a power grid load curve according to the power load P corresponding to the time point.

[0024] Specifically, the voltage corresponding to each time point includes the voltages corresponding to 1 to n unit times collected at intervals of unit time, the current corresponding to each time point includes the currents corresponding to 1 to n unit times collected at intervals of unit time, the power load corresponding to each time point includes the power loads corresponding to 1 to n unit times calculated according to the voltage corresponding to each time point and the current corresponding to each time point at intervals of unit time. The load power factor refers to the coefficient that power affects the power load. In this embodiment, the specific value of the power factor is not limited, and those skilled in the art can set it according to the actual situation as long as it meets the calculation requirements of the power load corresponding to each time point. For example, it can be set according to the types of power grid equipment. The power grid load curve refers to a curve graph made with time as the abscissa and the power load corresponding to each time point as the ordinate for intuitively showing the change of the power load.

[0025] Specifically, the power grid load curve analysis and prediction module is used to compare the power load Pi corresponding to each time point in the power grid load curve with the preset power load P0, judge the power load situation at this time point according to the comparison result, and predict the power grid load curve according to the judgment result, where: When P1 ≤ P0, the power grid load curve analysis and prediction module determines that the power load situation at this time point is in a normal state; When P1 > P0, the power grid load curve analysis and prediction module determines that the power load situation at this time point is in an abnormal state and predicts the power grid load curve.

[0026] Specifically, the power load Pi corresponding to each time point refers to the power load corresponding to each time point in the grid load curve. The preset power load refers to a preset value used to judge the power load situation at this time point. In this embodiment, the value of the preset power load is not limited, and those skilled in the art can freely set it according to needs, as long as it meets the judgment of the power load situation at the time point. For example, it can be set according to the types and quantities of various devices in the grid. The power load situation at this time point refers to the power load situation of the grid at this time point, and the power load situation at this time point includes that the power load situation at this time point is in a normal state and the power load situation at this time point is in an abnormal state.

[0027] Specifically, when the grid load curve analysis and prediction module predicts the grid load curve, the grid load curve analysis and prediction module selects a convolutional neural network model as the basic framework of the future load curve prediction model, and divides 75% of the historical grid load curve dataset into a prediction training set, 15% of the historical grid load curve dataset into a prediction validation set, and 10% of the historical grid load curve dataset into a prediction test set. The convolutional neural network model is trained according to the prediction training set, and the weights of the convolutional neural network model are updated through the backpropagation algorithm. After each epoch, the prediction validation set is used to verify the convolutional neural network model to obtain a prediction validation loss value and a prediction validation accuracy. When the validation loss value and the validation accuracy meet the preset prediction conditions, the convolutional neural network model is tested according to the prediction test set to obtain a prediction test accuracy. When the prediction test accuracy reaches the preset accuracy, the convolutional neural network model is output as the future load curve prediction model. The grid load curve analysis and prediction module converts the grid load curve into a picture form and inputs it into the future load curve prediction model, and obtains the future grid load curve output by the future load curve prediction model.

[0028] Specifically, the epoch refers to the process of a convolutional neural network model completing one forward propagation and one backward propagation on the prediction training set. The predicted loss value refers to the loss value of the loss function in the convolutional neural network model when the predicted validation set is input into the convolutional neural network model for verification. The verification accuracy rate refers to the ratio of the number of prediction results that are consistent with the prediction results in the predicted validation set to the total number of samples in the predicted validation set after the predicted validation set is input into the convolutional neural network model. The preset verification condition refers to that the verification loss value does not decrease and the verification accuracy rate does not increase for nine consecutive epochs. The predicted test accuracy rate refers to the ratio of the number of data test results that are consistent with the prediction results in the predicted test set to the total number of samples in the predicted test set after the predicted test set is input into the convolutional neural network model. The preset accuracy rate refers to the preset value of the predicted test accuracy rate for determining that the convolutional neural network model reaches the output standard. For example, the preset accuracy rate can be set to 95%. The historical power grid load curve library refers to the power grid load curves collected and produced historically, which are made into an atlas for training the future load curve prediction model. The real-time power grid load curve refers to the power grid load curve corresponding to the abnormal state of the power load situation at this time point determined by the power grid load curve analysis and prediction module. The future power grid load curve refers to the future change trend of the real-time power grid load curve predicted by the future load curve prediction model.

[0029] Specifically, the dispatching strategy system module compares the total power consumption load PR of each region in the future power grid load curve with the preset maximum power grid load PR0max and the preset minimum power grid load PR0min, judges the regional importance degree of this region according to the comparison result, and divides the region according to the judgment result, where: When PR≥PR0max, the regional division unit determines that the regional importance degree of this region is important and divides this region into an important load area; When PR0min<PR<PR0max, the regional division unit determines that the regional importance degree of this region is secondary and divides this region into an adjustable load area; When PR≤PR0min, the regional division unit determines that the regional importance degree of this region is unimportant and divides this region into a general load area.

[0030] Specifically, the total electricity load of each region refers to the sum of the electricity loads of each region within the power grid coverage. Each region within the power grid coverage refers to dividing the power grid coverage area into multiple regions with a circular range of 10 m radius as one region. The preset maximum power grid load refers to the maximum preset value used to judge the regional importance of this region, and the preset maximum power grid load also refers to the minimum preset value used to judge the regional importance of this region. The important load area refers to the area with extremely high requirements for the stability, continuity, and reliability of power supply. The adjustable load area refers to the area in the power system where the load equipment included in this region can flexibly change its power consumption or power consumption time within a certain range according to the operating state of the power grid and the power supply situation. The general load area refers to the area in the power system where the requirement for power supply reliability is relatively low, and the impact on production, life, etc. is small during power outages or power supply fluctuations. The regional importance of this region refers to the regional importance judged by comparing the total electricity load of each region with the preset maximum power grid load PR0max and the preset minimum power grid load PR0min, and the regional importance of this region includes that the regional importance of this region is important, the regional importance of this region is secondary, and the regional importance of this region is unimportant.

[0031] Specifically, the dispatching strategy system module performs power distribution according to the divided regions, where: For the important load area, according to the number of devices m in the important load area in the power grid load data and the rated power pj of the devices in the important load area, calculate the basic power distribution amount Pb of the important load area, and set , and calculate the actual power distribution amount Ps of the important load area according to Pb and the redundancy coefficient ω, and set Ps = ωPb, 1.1 < ω < 1.3; For the adjustable load area, according to the rated power pk of the devices in the adjustable load area in the power grid load data, the power grid load adjustment coefficient μ, and the device adjustment flexibility coefficient Fk, calculate the power distribution amount Pq of the adjustable load area, and set Pq = pk×(1 - μ×Fk); For the general load area, according to the actual power distribution amount Ps of the important load area, the power distribution amount Pq of the adjustable load area, and the total power supply amount Pt of the power grid, calculate the power supply amount Pf of the general load area, and set Pf = Pt - (Pq + Ps).

[0032] Specifically, the number of devices in the important load area refers to the quantity of power-consuming devices in the important load area. The rated power of the devices in the important load area refers to the rated power set for the devices in the important load area. The redundancy coefficient is a coefficient used to increase the redundancy of power distribution to ensure that the important load area can still operate normally under power demand fluctuations or emergencies. In this embodiment, the specific value of the redundancy coefficient is not limited, and those skilled in the art can set it according to requirements as long as it meets the calculation requirements for the actual power distribution amount in the important load area. For example, it can be set according to the number and type of devices in the important load area. The grid load adjustment coefficient is a coefficient for adjusting the calculated power distribution amount in the adjustable load area according to the overall load situation of the grid. In this embodiment, the specific value of the grid load adjustment coefficient is not limited, and those skilled in the art can set it according to requirements as long as it meets the calculation requirements for the power distribution amount in the adjustable load area. For example, it can be set according to the number and type of devices in the adjustable load area. The adjustment flexibility coefficient of the device is a coefficient indicating the degree to which its power consumption can be adjusted when the power demand changes. In this embodiment, the specific value of the adjustment flexibility coefficient of the device is not limited, and those skilled in the art can set it according to requirements as long as it meets the calculation requirements for the power distribution amount in the adjustable load area. For example, it can be set according to the number and type of devices in the adjustable load area. The total grid power supply refers to the total power that the grid can provide for all areas.

[0033] Specifically, the real-time monitoring and adjustment module compares the real-time total power Q of all areas in the grid load data with the preset power Q0, judges the power supply and demand situation of all areas based on the comparison result, and optimizes the power supply and demand situation of all areas according to the judgment result, where: When Q ≥ Q0, the real-time monitoring and adjustment module determines that the power supply and demand situation of all areas is sufficient for power supply and does not optimize the power supply and demand situation of all areas; When Q < Q0, the real-time monitoring and adjustment module determines that the power supply and demand situation of all areas is insufficient for power supply and optimizes the power supply and demand situation of all areas. The optimization method is to recalculate the actual power distribution amount in the important load area, the power distribution amount in the adjustable load area, and the power supply amount in the general load area according to the total grid power supply Pt, the first feedback coefficient f, the second feedback coefficient b, and the third feedback coefficient c, to obtain the adjusted actual power distribution amount Ps' in the important load area, the adjusted power distribution amount Pq' in the adjustable load area, and the adjusted power supply amount Pf' in the general load area, and set Ps' = f × Pt, Pq' = b × Pt, Ps' = c × Pt, f = 0.7, b = 0.3, c = 0.1.

[0034] Specifically, the total real-time power of each region refers to the real-time power value of each region. The preset power refers to a preset value used to judge the power supply and demand situation in this region. In this embodiment, the specific value of the preset power is not limited, and those skilled in the art can set it according to their needs, as long as it meets the judgment requirements for the power supply and demand situations in all regions. For example, it can be set according to the types and quantities of all electrical equipment in the power grid coverage area. The power supply and demand situation in all regions refers to the judgment of the power supply and demand in all regions based on the value of the total real-time power of each region after comparing the total real-time power of each region with the preset power. The power supply and demand situation in all regions includes that the power supply in all regions is sufficient and the power supply in all regions is insufficient.

[0035] Specifically, the real-time monitoring and adjustment module collects the voltages of each region in real time under the actual power distribution in the important load area after adjustment, the power distribution in the adjustable load area after adjustment, and the power supply in the general load area after adjustment, compares the total collected voltage UZ with the preset total voltage UZ0, and judges the equipment failure state according to the comparison result, where: When UZ ≤ UZ0, the real-time monitoring and adjustment module determines that the equipment failure state is normal; When UZ > UZ0, the real-time monitoring and adjustment module determines that the equipment failure state is abnormal.

[0036] Specifically, when the real-time monitoring and adjustment module determines that the equipment failure state is normal, it calculates the adjusted failure voltage Us2 of the important load area according to the adjusted voltage Us1 of the important load area and the first failure parameter C, sets Us2 = C × Us1, compares the adjusted failure voltage Us2 of the important load area with the preset first adjusted failure voltage Us0, judges the failure situation of the important load area according to the comparison result, and gives a failure alarm according to the judgment result, where: When Us2 ≤ Us0, the real-time monitoring and adjustment module determines that the failure situation of the important load area is normal and does not give a failure alarm; When Us2 > Us0, the real-time monitoring and adjustment module determines that the failure situation of the important load area is abnormal and gives a failure alarm; Specifically, when the real-time monitoring and adjustment module determines that the equipment failure state is normal, it calculates the adjusted failure voltage Ut2 of the adjustable load area according to the adjusted voltage Ut1 of the adjustable load area and the second failure parameter V, sets Ut2 = V × Ut1, compares the adjusted failure voltage Ut2 of the important load area with the preset second adjusted failure voltage Ut0, judges the failure situation of the adjustable load area according to the comparison result, and gives a failure alarm according to the judgment result, where: When Ut2 ≤ Ut0, the real-time monitoring and adjustment module determines that the fault condition of the adjustable load area is normal and does not perform a fault alarm. When Ut2 > Ut0, the real-time monitoring and adjustment module determines that the fault condition of the adjustable load area is abnormal and performs a fault alarm.

[0037] Specifically, when the real-time monitoring and adjustment module determines that the equipment fault state is normal, it calculates the adjusted fault voltage Uy2 of the general load area according to the adjusted voltage Uy1 of the general load area and the third fault parameter B, sets Uy2 = B × Uy1, compares the adjusted fault voltage Uy2 of the general load area with the preset third adjusted fault voltage Uy0, judges the fault condition of the general load area according to the comparison result, and performs a fault alarm according to the judgment result, where: When Uy2 ≤ Uy0, the real-time monitoring and adjustment module determines that the fault condition of the general load area is normal and does not perform a fault alarm. When Uy2 > Uy0, the real-time monitoring and adjustment module determines that the fault condition of the general load area is abnormal and performs a fault alarm.

[0038] Specifically, the preset total voltage refers to a preset value used to judge the fault state of the device. In this embodiment, the specific value of the preset total voltage is not limited. Those skilled in the art can set it according to their needs, as long as it meets the judgment requirements for the device fault state. For example, it can be set according to the types and quantities of all devices. The first fault parameter is the fault coefficient for calculating the adjusted fault voltage in the important load area. In this embodiment, the specific value of the first fault parameter is not limited. Those skilled in the art can set it according to their needs, as long as it meets the calculation requirements for the adjusted fault voltage in the important load area. For example, it can be set according to the types and quantities of the devices in the important load area. The preset first adjusted fault voltage refers to a preset value used to judge the fault condition in the important load area. In this embodiment, the specific value of the preset first adjusted fault voltage is not limited. Those skilled in the art can freely set it according to the actual needs, as long as it meets the judgment requirements for the fault condition in the important load area. For example, it can be set according to the types and quantities of the devices in the important load area. The second fault parameter is the fault coefficient for calculating the adjusted fault voltage in the adjustable load area in the important load area. In this embodiment, the specific value of the second fault parameter is not limited. Those skilled in the art can set it according to their needs, as long as it meets the calculation requirements for the adjusted fault voltage in the adjustable load area. For example, it can be set according to the types and quantities of the adjustable load area. The preset second adjusted fault voltage refers to a preset value used to judge the fault condition in the adjustable load area. In this embodiment, the specific value of the preset second adjusted fault voltage is not limited. Those skilled in the art can freely set it according to the actual needs, as long as it meets the judgment requirements for the fault condition in the adjustable load area. For example, it can be set according to the types and quantities of the devices in the adjustable load area. The third fault parameter is the fault coefficient for calculating the adjusted fault voltage in the general load area in the important load area. In this embodiment, the specific value of the third fault parameter is not limited. Those skilled in the art can set it according to their needs, as long as it meets the calculation requirements for the adjusted fault voltage in the general load area. For example, it can be set according to the types and quantities of the devices in the general load area. The preset third adjusted fault voltage refers to a preset value used to judge the fault condition in the general load area. In this embodiment, the specific value of the preset third adjusted fault voltage is not limited. Those skilled in the art can freely set it according to the actual needs, as long as it meets the judgment requirements for the fault condition in the general load area. For example, it can be set according to the types and quantities of the devices in the general load area.

[0039] Specifically, the real-time monitoring and adjustment module also calculates the device power factor cosλ according to the active power S and reactive power D of the device in the grid load data, and sets , compare cosλ with the preset power factor cosλ0, judge the reactive power compensation situation of the device according to the comparison result, and correct the judgment situation of the power supply and demand situation in the area according to the judgment result, where, When cosλ≥cosλ0, the real-time monitoring and adjustment module determines that the reactive power compensation situation of the device is sufficient and does not correct the judgment situation of the power supply and demand situation in the area; When cosλ<cosλ0, the real-time monitoring and adjustment module determines that the reactive power compensation situation of the device is insufficient and corrects the judgment situation of the power supply and demand situation in the area through the correction coefficient =0.8 + 0.2e -（cosλ0-cosλ） , cosλ>0, correct the preset power to obtain the corrected preset power Q0b, and set Q0b = ×Q0, and compare the corrected preset power Q0b with the real-time power Q of each area again, and re-judge the power supply and demand situation in the area according to the comparison result.

[0040] Specifically, the active power S of the device refers to the electric energy actually consumed by the device per unit time. The reactive power Q of the device refers to the power part that does not do work due to the phase difference between the current and the voltage due to the existence of inductance and capacitance in the AC circuit. The power factor cosλ is the ratio of the active power S to the apparent power, which reflects how much electric energy is converted into useful work when the device consumes electric energy, and how much electric energy circulates in the power grid in the form of reactive power. The apparent power refers to the product of voltage and current. The preset power factor refers to the preset value used to judge the reactive power compensation situation of the device. In this embodiment, the specific value of the preset power factor is not limited, and those skilled in the art can set it according to needs, as long as it meets the judgment requirements of the reactive power compensation situation of the device, for example, it can be set according to the type and parameters of the device.

[0041] Please refer to Figure 2 shown, which is a schematic flow chart of the power grid load optimization intelligent control method of this embodiment. The method includes, Step S1, collect the power grid load data in the power grid equipment; Step S2, make a power grid load curve according to the collected power grid load data, judge the abnormality of the power grid load curve, and also use the future load curve prediction model to predict the future trend of the abnormal power grid load curve to obtain the future power grid load curve; Step S3, divide the industrial electricity consumption area according to the total electricity consumption load of each area of the future power grid load curve to obtain the divided area; Step S4, monitor the real-time operating status of devices in different regions, and optimize the power distribution according to the real-time operating status; Step S5, calculate the power factor of the device, judge the reactive power compensation situation of the device according to the power factor of the device, and correct the optimization process of the power distribution according to the judgment result.

[0042] Specifically, by connecting the sensor with the data acquisition system, the load data of the power grid equipment can be collected in real time, ensuring the timeliness and accuracy of the data. Using the power grid load curve for anomaly judgment, the abnormal situations in the power grid operation can be detected in time, so as to take corresponding measures to avoid the occurrence of accidents. The industrial electricity consumption area is divided, and the power distribution is carried out according to the electricity consumption demands of different regions, realizing the refined management of power resources, which helps to optimize the power configuration, reduce energy waste, and improve the energy utilization efficiency. By monitoring the operating status of devices in different regions in real time and optimizing the power distribution according to the real-time data, the dynamic balance between power supply and electricity consumption demands can be ensured, which helps to improve the flexibility and response speed of the power system and meet the constantly changing electricity consumption demands. By reasonably distributing power resources, overloading operation of some regions or devices can be avoided, the service life of the devices can be extended, and the failure rate can be reduced.

[0043] So far, the technical solution of the present invention has been described in combination with the preferred embodiments shown in the drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the protection scope of the present invention.

Claims

1. A power grid load optimization intelligent control system, characterized in that: include: A power grid load data acquisition module is used to collect power grid load data in power grid equipment; A power grid load curve analysis and prediction module, used to prepare a power grid load curve according to the power grid load data, to make anomaly judgments according to the power grid load curve, and to predict the power grid load curve to obtain a future power grid load curve; A dispatching strategy system module is used to divide the industrial power consumption area according to the future power grid load curve, obtain the divided areas, and distribute power according to the divided areas; The real-time monitoring and adjustment module is used to monitor the real-time operating status of the equipment in each divided area, optimize the power distribution according to the operating status of the equipment in each divided area, and correct the optimization process of power distribution according to the power factor of the equipment.

2. The power grid load optimization intelligent control system according to claim 1 is characterized in that: The grid load curve analysis and prediction module calculates the corresponding power load P={p1, p2, p3...pn} at each time point according to the corresponding voltage U={u1, u2, u3...un} at each time point, the corresponding current A={a1, a2, a3...an} at each time point and the load power factor cosβ in the grid load data, sets P=UAcosβ, and makes a grid load curve according to the corresponding power load P at the time point.

3. The power grid load optimization intelligent control system according to claim 2 is characterized in that: The power grid load curve analysis and prediction module is used to compare the power load Pi corresponding to each time point in the power grid load curve with the preset power load P0, and judge the power load situation at the time point according to the comparison result, and predict the power grid load curve according to the judgment result, wherein: When P1≤P0, the power grid load curve analysis and prediction module determines that the power load situation at this time point is normal; When P1>P0, the power grid load curve analysis and prediction module determines that the power load situation at this time point is abnormal, and predicts the power grid load curve.

4. The power grid load optimization intelligent control system according to claim 3 is characterized in that: When the power grid load curve analysis and prediction module predicts the power grid load curve, the power grid load curve analysis and prediction module selects a convolutional neural network model as the basic framework of the future load curve prediction model, and divides 70% of the historical power grid load curve data set into a prediction training set, 15% of the historical power grid load curve data set into a prediction verification set, and 15% of the historical power grid load curve data set into a prediction test set. The convolutional neural network model is trained according to the prediction training set, and the weight of the convolutional neural network model is updated by a back propagation algorithm. After each epoch, the convolutional neural network model is verified using the prediction verification set to obtain a prediction verification loss value and a prediction verification accuracy rate. When the verification loss value and the verification accuracy rate meet the preset prediction conditions, the convolutional neural network model is tested according to the prediction test set to obtain a prediction test accuracy rate. When the prediction test accuracy rate reaches the preset accuracy rate, the convolutional neural network model is output as a future load curve prediction model. The power grid load curve analysis and prediction module converts the power grid load curve into a picture form, inputs it into the future load curve prediction model, and obtains the future power grid load curve output by the future load curve prediction model.

5. The power grid load optimization intelligent control system according to claim 4 is characterized in that: The dispatching strategy system module compares the total power load PR of each area in the future power grid load curve with the preset maximum power grid load PR0max and the preset minimum power grid load PR0min, judges the regional importance of the area according to the comparison result, and divides the area according to the judgment result, wherein: When PR≥PR0max, the area division unit determines that the area importance of the area is important, and divides the area into an important load area; When PR0min<PR<PR0max, the area division unit determines that the area importance of the area is secondary, and divides the area into an adjustable load area; When PR≤PR0min, the area division unit determines that the area importance of the area is not important, and divides the area into a general load area.

6. The power grid load optimization intelligent control system according to claim 5 is characterized in that: The dispatch strategy system module distributes electricity according to the divided areas, wherein: Important load area, according to the number of equipment m in the important load area in the power grid load data and the rated power pj of the equipment in the important load area, calculate the basic power distribution amount Pb of the important load area, set , and calculate the actual power distribution Ps of the important load area based on Pb and the redundancy coefficient ω, setting Ps=ωPb, 1.1<ω<1.3; The adjustable load area calculates the power distribution amount Pq of the adjustable load area according to the rated power pk of the adjustable load area equipment in the power grid load data, the power grid load adjustment coefficient μ and the adjustment flexibility coefficient Fk of the equipment, and sets Pq=pk×(1-μ×Fk); For general load areas, according to the actual power distribution Ps in important load areas, the power distribution Pq in the load area can be adjusted, the total power supply Pt of the power grid can be used to calculate the power supply Pf in the general load area, and set Pf=Pt-(Pq+Ps).

7. The power grid load optimization intelligent control system according to claim 6 is characterized in that: The real-time monitoring and adjustment module compares the real-time total power Q of all regions in the power grid load data with the preset power Q0, and judges the power supply and demand of all regions according to the comparison results, and optimizes the power supply and demand of all regions according to the judgment results, wherein: When Q≥Q0, the real-time monitoring and adjustment module determines that the power supply and demand situation in all regions is sufficient, and does not optimize the power supply and demand situation in all regions; When Q<Q0, the real-time monitoring and adjustment module determines that the power supply and demand situation in all regions is insufficient, and optimizes the power supply and demand situation in all regions. The optimization method is to recalculate the actual power distribution amount of the important load area, the power distribution amount of the adjustable load area and the power supply amount of the general load area according to the total power supply of the power grid Pt, the first feedback coefficient f, the second feedback coefficient b and the third feedback coefficient c, and obtain the actual power distribution amount Ps' of the important load area after adjustment, the power distribution amount Pq' of the adjustable load area after adjustment and the power supply amount Pf' of the general load area after adjustment, and set Ps'=f×Pt, Pq'=b×Pt, Ps'=c×Pt, f=0.7, b=0.3, c=0.

1.

8. The power grid load optimization intelligent control system according to claim 7 is characterized in that: The real-time monitoring and adjustment module collects the actual power distribution of the important load area after adjustment, the power distribution of the adjustable load area after adjustment, and the voltage of each area under the power supply of the general load area after adjustment in real time, compares the collected total voltage UZ with the preset total voltage UZ0, and judges the equipment fault state according to the comparison result, wherein: When UZ≤UZ0, the real-time monitoring and adjustment module determines that the equipment fault state is normal; When UZ>UZ0, the real-time monitoring and adjustment module determines that the equipment fault state is abnormal.

9. The power grid load optimization intelligent control system according to claim 8, characterized in that: The real-time monitoring and adjustment module also calculates the equipment power factor cosλ according to the equipment active power S and equipment reactive power D in the power grid load data, and sets , and compare cosλ with the preset power factor cosλ0, judge the reactive power compensation of the equipment according to the comparison result, and correct the judgment of the power supply and demand situation in the area according to the judgment result, wherein, When cosλ≥cosλ0, the real-time monitoring and adjustment module determines that the reactive power compensation of the equipment is sufficient, and does not modify the judgment of the power supply and demand situation in the area; When cosλ<cosλ0, the real-time monitoring and adjustment module determines that the reactive power compensation of the equipment is insufficient, and corrects the judgment of the power supply and demand situation in the area, and uses the correction coefficient =0.8+0.2e -（cosλ0-cosλ） , cosλ>0, correct the preset power, get the corrected preset power Q0b, set Q0b= ×Q0, and re-compare the revised preset power Q0b with the real-time power Q of each area, and re-judge the power supply and demand situation of the area based on the comparison results.

10. The intelligent control system for optimizing power grid load according to claims 1-9, characterized in that: Also included is a control method of the system, the method comprising: Step S1, collecting grid load data in grid equipment; Step S2, making a power grid load curve according to the collected power grid load data, and making an abnormal judgment on the power grid load curve, and also using a future load curve prediction model to predict the future trend of the abnormal power grid load curve to obtain a future power grid load curve; Step S3, dividing the industrial power consumption area into regions according to the total power consumption load of each region of the future power grid load curve to obtain divided regions; Step S4, monitoring the real-time operating status of equipment in different areas, and optimizing power distribution according to the real-time operating status; Step S5, calculating the equipment power factor, judging the equipment reactive power compensation situation according to the equipment power factor, and correcting the optimization process of power distribution according to the judgment result.

Images