A method for reactive power optimization of power supply network in steel enterprises

By constructing interface interaction channels and improving genetic algorithms in the power grid of steel enterprises, the problem of reactive power optimization relying on experience was solved, network losses were reduced and power quality was improved, ensuring the safe and stable operation of the system.

CN123074009APending Publication Date: 2026-11-10BAOSTEEL ZHANJIANG IRON & STEEL CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611518701.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-29
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing reactive power optimization techniques rely on manual experience in the power systems of steel enterprises, resulting in high network losses and large errors in optimization results, without fully considering the actual reactive power compensation constraints.

Method used

By establishing an interface interaction channel, real-time data is uploaded to the reactive power optimization monitoring platform, a reactive power optimization model is constructed, and an improved genetic algorithm is used to solve the model. Taking into account the actual constraints of reactive power compensation in the power grid of steel enterprises, the switching status of equipment and transformer taps are optimized.

Benefits of technology

This has resulted in reduced network losses, improved power quality, increased economic efficiency, and ensured the safe and stable operation of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123074009A_ABST
    Figure CN123074009A_ABST
Patent Text Reader

Abstract

This invention discloses a reactive power optimization method for a power supply network in a steel enterprise. Data from various substation areas is uploaded to a reactive power optimization monitoring platform built on CyberControl. Simultaneously, an interface interaction channel is established between MATLAB and CyberControl to enable model data exchange and file format transmission. A reactive power optimization model for the steel enterprise power grid is established, including the objective function, power flow calculation equations, and reactive power optimization constraints. An improved genetic algorithm is used on the MATLAB platform to solve the reactive power optimization model, calculating the reactive power optimization results for the steel enterprise power supply network. The optimization results are then sent to the reactive power optimization monitoring platform for display. Using this method, the optimal reactive power optimization scheme can be obtained, thereby reducing network losses, improving power quality, increasing economic efficiency, and ensuring the safe and stable operation of the system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of reactive power optimization calculation for power grids, specifically involving a reactive power optimization method for power supply networks of steel enterprises that considers the actual constraints of reactive power compensation. Background Technology

[0002] Steel enterprises are major energy consumers in my country, with electricity being one of the most important energy sources after coal in steel production. Reactive power optimization plays a crucial role in energy conservation and emission reduction for the entire enterprise and is an important measure to ensure reliable system operation. Rational planning and allocation of existing reactive power resources in the power grid is of great significance for improving power quality and increasing economic efficiency. Insufficient or improperly allocated reactive power in the power system can lead to a series of problems, including reduced voltage levels, increased losses, decreased system stability, and malfunctions of user equipment, potentially even causing system collapse. Generally, reactive power optimization research includes two key steps: establishing a reactive power optimization model and determining the optimization algorithm. The key to establishing a reactive power optimization model is determining the objective function and constraints. Optimization algorithms include traditional algorithms such as gradient algorithms, Newton's method, quadratic programming, and linear programming; and artificial intelligence algorithms such as genetic algorithms, particle swarm optimization, and simulated annealing.

[0003] Currently, most existing reactive power optimization technologies are designed for large power grids. Research on reactive power optimization in the power systems of steel enterprises is still in its early stages. It mainly relies on the experience of staff to arrange the output and switching of reactive power compensation equipment, resulting in high network losses. At the same time, existing reactive power optimization technologies do not collect real-time load voltage data from the field to communicate with the established reactive power optimization model, and do not fully consider the actual reactive power compensation constraints, resulting in large calculation errors in the optimization results. Summary of the Invention

[0004] The purpose of this invention is to provide a reactive power optimization method for the power supply network of steel enterprises. This method fully considers the actual constraints of reactive power compensation in the power grid of steel enterprises, obtains the optimal reactive power optimization scheme, thereby reducing network losses, improving power quality, increasing economic efficiency, and ensuring the safe and stable operation of the system.

[0005] To achieve the above objectives, the present invention adopts the following technical solution.

[0006] A method for optimizing reactive power in the power supply network of a steel enterprise, characterized by comprising the following steps: S1. Upload the data from each substation area to the reactive power optimization monitoring platform built by CyberControl using the IEC 104 protocol. At the same time, establish an interface interaction channel between MATLAB and CyberControl to realize model data interaction and file format transmission. S2. Based on the collected real-time data, establish a set of power flow calculation equations for the steel enterprise power grid; based on the configuration parameters of the reactive power compensation equipment of the steel enterprise power grid, determine the reactive power optimization constraints of the steel enterprise power grid; construct an objective function with the goal of minimizing the overall grid loss of the plant; and form a reactive power optimization model for the steel enterprise power grid that includes the objective function, the set of power flow calculation equations, and the reactive power optimization constraints. S3. On the MATLAB platform, an improved genetic algorithm is used to solve the reactive power optimization model and calculate the reactive power optimization results of the power supply network of steel enterprises. S4. Send the reactive power optimization results to the reactive power optimization monitoring platform for display.

[0007] Furthermore, in step 1, the specific implementation process of uploading the data from each substation area to the reactive power optimization monitoring platform built by CyberControl using the IEC104 protocol is as follows: Access the data required by the power monitoring system within the local substation, and ensure that the data on the control center server is consistent with the data at the local substation; Data acquisition is mainly achieved by uploading data through remote control units in the substation, using the IEC104 protocol, and then uploading it to the network cabinet in the control center's computer room via the industrial control network. Two application servers are installed on the power dispatching console. A configured backbone switch is used to network the various stations. The servers are connected to perform centralized data processing and access the CyberControl platform.

[0008] Furthermore, in step 1, an interface interaction channel between MATLAB and CyberControl is established, and the project files in CyberControl are converted into a CSV file containing the node power and voltage required for reactive power optimization. This file is then saved in the folder where the MATLAB program is located, and the file is read in the MATLAB platform for subsequent calculations.

[0009] Furthermore, step 2 uses minimizing the overall network loss as the objective function, expressed as: ; In the formula, This represents the total number of branches in the network. branch road , The electrical conductance between them; They are nodes , The voltage; They are nodes , The phase angle.

[0010] Furthermore, the power flow calculation equations are equality constraints, expressed as follows: ; In the formula, It is a node The generator set injects active power. It is a node The active power of the load, , It is a node , voltage, It is a node , Mutual conductance between them It is a node , Mutual susceptance between them It is a node The generator set injects reactive power. It is a node The reactive power of the load, It is a node Parallel capacitors compensate for reactive power.

[0011] Furthermore, the inequality constraints include control variable constraints and state variable constraints.

[0012] The control variable constraints include: acquiring data from on-site callable reactive power regulation equipment and constructing upper and lower limit constraints for the capacity of parallel capacitors, SVCs and SVGs, constructing upper and lower limit constraints for adjustable transformer taps, and constructing upper and lower limit constraints for the reactive power regulation capacity of self-provided generator sets and waste heat generator sets. The state variable constraints include: constructing upper and lower limits for grid node voltage and upper and lower limits for power factor based on the requirements of steel enterprises' production processes for grid voltage and power factor.

[0013] Furthermore, in step 3, an improved genetic algorithm is called on the MATLAB platform and combined with the collected actual field data to solve the reactive power optimization model, so as to obtain the switching status of reactive power compensation equipment, the tap position of adjustable transformer, and the reactive power capacity of self-provided generator sets and waste heat and waste energy generator sets, so as to complete the reactive power optimization of the power supply network of steel enterprises.

[0014] Furthermore, in step S3, the specific process of solving the reactive power optimization model using an improved genetic algorithm includes: Read in parameters, Initial power flow calculation, Generate the initial population, Perform power flow calculations for each individual. Calculate the fitness function value. Individual choice, Adaptive crossover mutation is used until the convergence condition is met, at which point the optimal reactive power optimization result is output.

[0015] Furthermore, in step S4, the specific process of sending the reactive power optimization results to the reactive power optimization monitoring platform for display is as follows: the MATLAB platform outputs a CSV file containing the reactive power optimization results, the CyberControl platform reads the CSV file, and displays it on the reactive power optimization monitoring platform interface. The displayed content includes reactive power optimization suggestions for callable reactive power compensation devices, the reduction in network loss after optimization compared to before optimization, the power saving after optimization compared to before optimization, and the power cost saving after optimization compared to before optimization.

[0016] Compared with related technologies, the present invention has the following beneficial effects: This invention considers the actual constraints of reactive power compensation in steel enterprises. The data required for power flow equality constraints is transmitted to the MATLAB platform in real time via communication. Inequality constraints are acquired by accessing data from reactive power compensation equipment on-site, ensuring high reliability and good optimization results. The method is simple to operate, displaying optimization results such as network loss reduction and power savings on the reactive power monitoring platform interface, ensuring economic benefits and providing optimization suggestions to operators, making it convenient to use. Attached Figure Description

[0017] Figure 1 This is a flowchart of the reactive power optimization implementation technology for power supply networks of steel enterprises, which takes into account the actual constraints of reactive power compensation, provided by the present invention.

[0018] Figure 2 This is a diagram of the primary power supply and distribution system of Zhanjiang Iron and Steel Company.

[0019] Figure 3 This is the access topology diagram of the Zhanjiang Iron and Steel reactive power optimization monitoring platform system provided by the present invention.

[0020] Figure 4 This is a flowchart of the improved genetic algorithm provided by the present invention applied to reactive power optimization. Detailed Implementation

[0021] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0022] As shown in Figure 1, the present invention provides a method for optimizing reactive power in the power supply network of steel enterprises, comprising the following steps.

[0023] Step 1: Upload data from each substation area to the reactive power optimization monitoring platform built with CyberControl using the IEC 104 protocol. Simultaneously, establish an interface between MATLAB and CyberControl to enable model data exchange and file format transfer. The specific implementation process is as follows: Access the data required by the local substation's power monitoring system to ensure consistency between the data on the control center server and the local substation. Data acquisition primarily involves uploading data from remote actuators within the substation using the IEC104 communication protocol, and then transmitting it via the industrial control network to the network cabinet in the control center's computer room.

[0024] Two application servers are installed on the power dispatching console. A main switch is configured to network six substations (Baoneng, pelletizing, central, raw materials, coking, and steelmaking). The servers are connected to perform centralized data processing and access the CyberControl platform.

[0025] Create a CSV file from the project files in CyberControl containing the node power and voltage required for reactive power optimization, save it in the folder where the MATLAB program is located, and then read the file in the MATLAB platform for subsequent calculations.

[0026] Step 2: Based on the collected real-time data, establish a set of power flow calculation equations for the steel enterprise power grid; determine the reactive power optimization constraints for the steel enterprise power grid based on the configuration parameters of the reactive power compensation equipment; construct an objective function with the goal of minimizing the overall grid loss; and form a reactive power optimization model for the steel enterprise power grid that includes the objective function, the set of power flow calculation equations, and the reactive power optimization constraints.

[0027] like Figure 2 As shown, the Zhanjiang Iron and Steel power grid hub substation includes the 220 / 110kV Baoneng Substation and the 110kV Central Substation. The 110kV downstream step-down substations mainly include: raw material substation, coke oven substation, steelmaking substation, pelletizing substation, and also smaller substations such as thick plate substation, oxygen production substation, hot rolling substation, cold rolling substation, and blast furnace blast station. The objective function is to minimize the overall network loss, expressed as: ; In the formula, This represents the total number of network branches, which is 54. branch road , The electrical conductance between them; They are nodes , The voltage; They are nodes , The phase angle.

[0028] Among them, the power flow calculation equations for steel enterprises are based on equality constraints, and the expressions are as follows: ; In the formula, It is a node The generator set injects active power. It is a node The active power of the load, , It is a node , voltage, It is a node , Mutual conductance between them It is a node , Mutual susceptance between them It is a node The generator set injects reactive power. It is a node The reactive power of the load, It is a node The parallel capacitors compensate for reactive power. The active power injected by the generator set, the active power of the load, the reactive power of the load, and the reactive power of the parallel capacitors, SVCs, and SVG reactive power compensation devices are obtained by reading the CSV file transmitted by the CyberControl platform.

[0029] The reactive power optimization constraints correspond to the inequality constraints of the reactive power compensation equipment that can be called on site, including control variable constraints and state variable constraints.

[0030] Control variable constraints: Obtain data from available reactive power regulation equipment on-site, and construct upper and lower limit constraints for the capacity of parallel capacitors, SVCs, and SVGs; upper and lower limit constraints for adjustable transformer tap changes; and upper and lower limit constraints for the reactive power regulation capacity of existing generator sets (including self-provided generator sets and waste heat generator sets). Specifically, such as... Figure 2 In the power grid, the available reactive power compensation equipment mainly includes parallel capacitors, SVCs, SVGs, etc., totaling 93 units, connected to 16 nodes on the 10kV side. There is also one adjustable transformer. Existing generator sets include 4 self-provided generator sets and 4 waste heat and waste energy generator sets (including 2 waste heat and waste pressure TRT units and 2 sintering generator sets), connected to 8 nodes. The specific constraint equations are expressed as follows: ; In the formula, For nodes The capacity of the reactive power compensation equipment is a discrete variable. They are nodes Upper and lower limits of reactive power compensation equipment capacity. For nodes The reactive power capacity of the connected generator sets is a continuous variable. node The connected generator set has adjustable upper and lower limits for reactive power capacity. This represents the tap position of an adjustable transformer and is a discrete variable. This represents the upper and lower limits of the adjustable range of the transformer.

[0031] State variable constraints: Based on the requirements of steel enterprises' production processes for grid voltage and power factor, upper and lower limits of grid node voltage and power factor constraints are constructed. Each substation is required to have a bus voltage range not exceeding 5% above or below the rated voltage, and a power factor of not less than 0.9 for important nodes. The specific constraint equations are as follows: ; In the formula, Representative node voltage, Represents the rated voltage. Representing key nodes The power factor.

[0032] Step 3: On the MATLAB platform, an improved genetic algorithm is used to solve the reactive power optimization model, and the reactive power optimization results of the power supply network of the steel enterprise are calculated. Specifically, by calling the improved genetic algorithm and collecting actual field data, the optimal results of the switching status of reactive power compensation equipment, the tap position of adjustable transformers, and the reactive power capacity of existing generator sets (including self-provided generator sets and waste heat and waste energy generator sets) are obtained. Among them, in the power grid shown in Figure 2, the optimal switching capacity of reactive power compensation equipment at 16 nodes, the optimal reactive power regulation capacity of generator sets at 8 nodes, and the optimal tap position of 1 adjustable transformer are obtained.

[0033] The improved genetic algorithm is used to solve the reactive power optimization model, as shown in Figure 4, which is a flowchart of the application of the improved genetic algorithm provided by this invention to reactive power optimization. Specific steps include: reading in parameters, initial power flow calculation, generating an initial population, performing power flow calculation for each individual, calculating the fitness function value, individual selection, adaptive crossover and mutation, and outputting the optimal reactive power optimization result when the convergence condition is met.

[0034] Step 4: Return the optimization results to the reactive power optimization monitoring platform for display.

[0035] The MATLAB platform outputs a CSV file containing the reactive power optimization results. The CyberControl platform reads the CSV file and displays it on the reactive power optimization monitoring platform interface. The displayed content mainly includes: the reduction in network loss after optimization compared to before optimization, the amount of electricity saved after optimization compared to before optimization, and the amount of electricity cost saved after optimization compared to before optimization.

[0036] By employing the method of this invention, and fully considering the actual constraints of reactive power compensation in the power grid of steel enterprises, the optimal solution for reactive power optimization can be obtained, thereby reducing network losses, improving power quality, increasing economic efficiency, and ensuring the safe and stable operation of the system.

[0037] Those skilled in the art should recognize that the above examples are merely illustrative of the invention and are not intended to limit the scope of the invention. Any variations or modifications to the above embodiments that are within the scope of the invention's implementation will fall within the scope of the invention's claims.

Claims

1. A method for optimizing reactive power in the power supply network of an iron and steel enterprise, characterized in that, Includes the following steps: S1. Upload the data from each substation area to the reactive power optimization monitoring platform built by CyberControl using the IEC 104 protocol. At the same time, establish an interface interaction channel between MATLAB and CyberControl to realize model data interaction and file format transmission. S2. Based on the collected real-time data, establish a set of power flow calculation equations for the steel enterprise power grid; based on the configuration parameters of the reactive power compensation equipment of the steel enterprise power grid, determine the reactive power optimization constraints of the steel enterprise power grid; construct an objective function with the goal of minimizing the overall grid loss of the plant; and form a reactive power optimization model for the steel enterprise power grid that includes the objective function, the set of power flow calculation equations, and the reactive power optimization constraints. S3. On the MATLAB platform, an improved genetic algorithm is used to solve the reactive power optimization model and calculate the reactive power optimization results of the power supply network of steel enterprises. S4. Send the reactive power optimization results to the reactive power optimization monitoring platform for display.

2. The reactive power optimization method for the power supply network of steel enterprises according to claim 1, characterized in that, In step 1, the specific implementation process of uploading the data from each substation area to the reactive power optimization monitoring platform built by CyberControl using the IEC104 protocol is as follows: Access the data required by the power monitoring system within the local substation, and ensure that the data on the control center server is consistent with the data at the local substation; Data acquisition is mainly achieved by uploading data through remote control units in the substation, using the IEC104 protocol, and then uploading it to the network cabinet in the control center's computer room via the industrial control network. Two application servers are installed on the power dispatching console. A configured backbone switch is used to network the various stations. The servers are connected to perform centralized data processing and access the CyberControl platform.

3. The reactive power optimization method for the power supply network of steel enterprises according to claim 1, characterized in that, In step 1, an interface interaction channel between MATLAB and CyberControl is established. The project files in CyberControl are converted into a CSV file containing the node power and voltage required for reactive power optimization. This file is then saved in the folder where the MATLAB program is located. The MATLAB platform reads this file for subsequent calculations.

4. The reactive power optimization method for the power supply network of steel enterprises according to claim 1, characterized in that, In step 2, the objective function expression for minimizing the overall network loss of the plant is constructed as follows: ; In the formula, This represents the total number of branches in the network. branch road , The electrical conductance between them; They are nodes , The voltage; They are nodes , The phase angle.

5. The reactive power optimization method for the power supply network of steel enterprises according to claim 1, characterized in that, The power flow calculation equations are equality constraints, expressed as follows: ; In the formula, It is a node The generator set injects active power. It is a node The active power of the load, , It is a node , voltage, It is a node , Mutual conductance between them It is a node , Mutual susceptance between them It is a node The generator set injects reactive power. It is a node The reactive power of the load, It is a node Parallel capacitors compensate for reactive power.

6. The reactive power optimization method for the power supply network of steel enterprises according to claim 1, characterized in that, The reactive power optimization constraints are inequality constraints, including control variable constraints and state variable constraints.

7. The reactive power optimization method for the power supply network of steel enterprises according to claim 6, characterized in that, The control variable constraints include: based on the configuration parameters of the power grid reactive power compensation equipment, constructing upper and lower limit constraints for the capacity of parallel capacitors, SVCs and SVGs, upper and lower limit constraints for adjustable transformer taps, and upper and lower limit constraints for the reactive power regulation capacity of self-provided generator sets and waste heat generator sets; the state variable constraints include: based on the requirements of steel enterprises' production processes for power grid voltage and power factor, constructing upper and lower limit constraints for power grid node voltage and power factor.

8. The reactive power optimization method for the power supply network of steel enterprises according to claim 1, characterized in that, In step 3, an improved genetic algorithm is used on the MATLAB platform, combined with the collected actual field data, to solve the reactive power optimization model. This yields the switching status of the reactive power compensation equipment, the tap position of the adjustable transformer, and the reactive power capacity of the self-provided generator set and the waste heat and waste energy generator set, thereby completing the reactive power optimization of the power supply network of the steel enterprise.

9. The reactive power optimization method for the power supply network of steel enterprises according to claim 8, characterized in that, The specific process of solving the reactive power optimization model using an improved genetic algorithm includes: Read in parameters, Initial power flow calculation, Generate the initial population, Perform power flow calculations for each individual. Calculate the fitness function value. Individual choice, Adaptive crossover mutation is used until the convergence condition is met, at which point the optimal reactive power optimization result is output.

10. The reactive power optimization method for the power supply network of steel enterprises as described in claim 1, characterized in that, In step S4, the specific process of sending the reactive power optimization results to the reactive power optimization monitoring platform for display is as follows: the MATLAB platform outputs a CSV file containing the reactive power optimization results, the CyberControl platform reads the CSV file, and displays it on the reactive power optimization monitoring platform interface. The displayed content includes reactive power optimization suggestions for callable reactive power compensation devices, the amount of network loss reduction after optimization compared to before optimization, the amount of electricity saved after optimization compared to before optimization, and the amount of electricity cost saved after optimization compared to before optimization.