Network source load storage optimization method and device for low-voltage transformer area

Abstract

The invention discloses a grid source load storage optimization method for a low-voltage transformer area, and the method comprises the steps: obtaining power grid resource data and historical load data of the low-voltage transformer area, carrying out the investigation and evaluation of the power grid resource data, and carrying out the preprocessing and feature extraction of the historical load data; constructing a load prediction model, and training the load prediction model by using historical load data; the load demand change of the low-voltage transformer area is predicted, and load prediction data is acquired and monitored; the peak-valley difference value needing to be adjusted and the total energy storage amount of the energy storage equipment are calculated, and the energy storage equipment is deployed; performing real-time optimization and supervision on the network source load storage system by using an intelligent control algorithm to obtain real-time data; and obtaining an optimal resource configuration optimization scheme of the network source load storage system, and performing power grid resource scheduling. According to the method, the real-time data and the prediction model are utilized, power grid resource allocation is optimized, power balance and operation cost minimization are achieved, the operation efficiency of the power grid is improved, energy consumption and emission are reduced, and contribution is made to sustainable energy development.

Description

Technical Field

[0001] The present invention relates to the field of optimization technology, and in particular to a network source, load and storage optimization method and device for a low-voltage station area. Background Art

[0002] As my country's electricity demand continues to increase, especially in low-voltage areas, the power supply quality and stability have a greater impact on users due to their wide distribution, low load density, and large power supply radius; low-voltage area optimization has become an important aspect of power system operation and management.

[0003] The grid source, load and storage of low-voltage substations is a comprehensive concept, which involves the optimal configuration and coordinated operation of the four key components of the low-voltage power system: power grid, power supply, load and energy storage. The traditional low-voltage substation grid source, load and storage optimization method is mainly based on experience and rules, but it still has some shortcomings in actual use. For example, historical data analysis requires a large amount of reliable data, and many low-voltage substations lack a complete data collection and management system; it can often only optimize local problems and it is difficult to achieve the global optimal grid source, load and storage configuration.

[0004] In summary, it is necessary to design a grid source, load and storage optimization method for low-voltage substations to optimize grid resource allocation, achieve power balance and minimize operating costs. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide a network source load storage optimization method for a low voltage substation to solve the technical problem that the prior art has no perfect data collection and management system and can only optimize local problems.

[0006] In a first aspect, an embodiment of the present invention provides a method for optimizing network source, load and storage in a low voltage area, comprising:

[0007] S1. Obtain the power grid resource data and historical load data of the low-voltage substation, investigate and evaluate the power grid resource data, and pre-process and extract features of the historical load data;

[0008] S2. Based on the processed power grid resource data and historical load data, a load forecasting model is constructed, and the load forecasting model is trained using the historical load data;

[0009] S3. Based on the trained load forecasting model, predict the load demand changes in the low-voltage area, and obtain and monitor the load forecasting data;

[0010] S4. Calculate the peak-to-valley difference and total energy storage required to be adjusted by the energy storage device based on load forecast data and grid stability requirements, and deploy the energy storage device;

[0011] S5. Use intelligent control algorithms to optimize and monitor the grid-source-load-storage system in real time and obtain real-time data;

[0012] S6. Combine real-time data and load forecast data to obtain the optimal resource configuration optimization plan for the grid source, load and storage system, and perform grid resource dispatch based on the obtained optimal resource configuration optimization plan.

[0013] Preferably, the power grid resource data includes power supply capacity, power load, current status of distribution network, and energy structure.

[0014] Preferably, the historical load data includes load changes in different time periods and under different circumstances.

[0015] Preferably, S2 also includes using a cross-validation method to evaluate and optimize the load forecasting model.

[0016] Preferably, S4 also includes performing a technical evaluation of the energy storage technology based on the load forecast data and the grid stability requirements, determining the maximum load and the minimum load in the selected time period, and calculating the peak-to-valley difference that the energy storage device needs to adjust;

[0017] Calculate the energy difference within the selected time period based on the renewable energy generation forecast data and load demand forecast data within the selected time period;

[0018] determining a total energy storage amount required by the energy storage device based on the energy difference;

[0019] Combined with the grid architecture and load distribution, energy storage systems are centrally deployed in substations and distribution stations for centralized control and scheduling.

[0020] The calculation formula of the peak-to-valley difference is: ΔB = B max -B min ;

[0021] In the formula, ΔB is the peak-to-valley difference, that is, the power difference that needs to be adjusted by the energy storage device; B max is the maximum load in the selected time period, B min is the minimum load in the selected time period

[0022] The energy difference is calculated as: ΔC(t) = C lod (t)-C res (t);

[0023] Where ΔC(t) is the energy difference within time t, C lod (t) is the predicted amount of renewable energy power generation within time t, C res (t) is the load demand forecast within time t;

[0024] The total energy storage is calculated as: D总 =∑ t |ΔC(t)|;

[0025] Where D 总 is the total energy storage required by the energy storage device, and |ΔC(t)| is the absolute value of the energy difference within time t.

[0026] Preferably, the RL algorithm and Internet of Things technology are used to monitor the status of the grid-source-load-storage system in real time to detect potential faults and anomalies in a timely manner.

[0027] Preferably, the load forecasting model is calculated as:

[0028] Y(t)=a0+a1X1(t)+……+a n X n (t)+α(t),

[0029] Where Y(t) represents the predicted value of power load within time t, X i (t) is the observed value of the i-th influencing factor at time t, a i is the regression coefficient of the ith influencing factor, n is the total number of influencing factors, α(t) is the random interference term, α(t)~N(0, β 2 ), β 2 Expressed as the variance of the error term.

[0030] In a second aspect, a network source load storage optimization device for a low voltage station area includes:

[0031] Data processing module: used to obtain the power grid resource data and historical load data of the low-voltage substation, investigate and evaluate the power grid resource data, and pre-process and extract features of the historical load data;

[0032] Model building module: used to build a load forecasting model based on the processed power grid resource data and historical load data, and use the historical load data to train the load forecasting model;

[0033] Load forecast data acquisition module: used to predict the load demand changes in the low-voltage area based on the trained load forecast model, and to obtain and monitor load forecast data;

[0034] Energy storage equipment deployment module: used to calculate the peak-to-valley difference and total energy storage capacity that the energy storage equipment needs to adjust according to load forecast data and grid stability requirements, and deploy energy storage equipment;

[0035] Real-time data acquisition module: used to optimize and monitor the grid-source, load-storage system in real time using intelligent control algorithms and acquire real-time data;

[0036] Optimization scheme design module: used to combine real-time data and load forecast data to obtain the optimal resource configuration optimization plan for the grid source, load and storage system, and perform grid resource dispatch based on the obtained optimal resource configuration optimization plan.

[0037] The beneficial effects of the present invention are as follows:

[0038] The present invention optimizes the allocation of grid resources by making full use of real-time data and prediction models, achieves power balance and minimizes operating costs, can improve grid operation efficiency, reduce energy consumption and emissions, and contribute to sustainable energy development. It can also identify potential fault hazards, conduct early warning and maintenance in a timely manner, and reduce the failure rate.

[0039] Other features and advantages of the present invention will be described in the following description, and some features will become apparent from the description or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0040] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 A structural schematic diagram of a network source, load and storage optimization method for a low voltage station area provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] To facilitate understanding of this embodiment, Figure 1 A network source, load and storage optimization method for a low-voltage station area disclosed in an embodiment of the present invention is introduced in detail.

[0045] Embodiment 1: The embodiment of the present invention provides a method for optimizing network source, load and storage in a low-voltage area, including:

[0046] S1. Obtain the power grid resource data and historical load data of the low-voltage substation, investigate and evaluate the power grid resource data, and pre-process and extract features of the historical load data;

[0047] The historical load data includes load changes in different time periods and under different circumstances, that is, load data under different circumstances such as weekdays, weekends and holidays; at the same time, the influence of some external factors, such as weather conditions, economic activities, etc., needs to be considered. Data preprocessing and feature extraction include converting date, time point, and holiday information into features that can be used for modeling.

[0048] The power grid resource data includes power supply capacity, power load, distribution network status, and energy structure. The power supply capacity includes power source type, power generation scale, line capacity, and substation capacity; the power load includes peak load, flat load, daily variation pattern, seasonal variation pattern, etc., analyze the structure of power load, and understand the power demand of different industries and different user groups; the distribution network status includes line length, line type, transformer capacity, and the type and number of distribution devices; the energy structure includes the proportion and share of electricity, natural gas, photovoltaic, wind power and other energy sources.

[0049] Furthermore, evaluate the balance of power supply and demand in low-voltage areas, analyze future supply and demand situations, and predict future power demand growth trends; investigate and evaluate the network loss rate of low-voltage power grids, analyze the causes of network losses and put forward improvement suggestions; evaluate the reliability of low-voltage power grids, including indicators such as failure rate and power outage time, and put forward measures to enhance reliability; evaluate the economic feasibility of low-voltage power grids, including operating costs, investment costs, returns and other indicators, and analyze the economic benefits under different development plans.

[0050] Furthermore, the methods for investigating the resources of the low-voltage power grid are as follows: Questionnaire survey: Conduct a questionnaire survey on electricity users to understand their electricity demand, electricity usage habits and other information; On-site inspection: Conduct an on-site inspection of the distribution network facilities in the low-voltage area to collect relevant data and information; Data analysis: Use existing power data, weather data, economic data, etc. to analyze and reveal the laws and trends of low-voltage power grid resources. By investigating and evaluating the resources of the low-voltage power grid, it can provide a decision-making basis for power grid planning, transformation, investment and other aspects, and help the low-voltage power grid to achieve safe, reliable, economical and environmentally friendly operation.

[0051] S2. Based on the processed power grid resource data and historical load data, a load forecasting model is constructed, and the load forecasting model is trained using the historical load data;

[0052] Specifically, a neural network model is selected to model the data; the selected model is trained using historical load data and the model parameters are adjusted to achieve the best effect; and the model is evaluated and optimized using the cross-validation method.

[0053] The load forecasting model is as follows:

[0054] Y(t)=a0+a1X1(t)+……+a n X n (t)+α(t),

[0055] Where Y(t) represents the predicted value of power load within time t, X i (t) is the observed value of the i-th influencing factor at time t, a i is the regression coefficient of the ith influencing factor, n is the total number of influencing factors, α(t) is the random interference term, α(t)~N(0, β 2 ), β 2 Expressed as the variance of the error term.

[0056] S3. Based on the trained load forecasting model, predict the load demand changes in the low-voltage area, and obtain and monitor the load forecasting data;

[0057] Specifically, the trained model is used to predict the future low-voltage area load; load data is monitored in real time, and the model is continuously optimized to improve accuracy. The prediction results are applied to the actual power system management to guide power system operators to reasonably arrange power generation plans, effectively adjust energy supply, and ensure the stable operation of the power grid.

[0058] S4. Calculate the peak-to-valley difference and total energy storage required to be adjusted by the energy storage device based on load forecast data and grid stability requirements, and deploy the energy storage device; specifically,

[0059] S41. Conduct a technical evaluation of energy storage technology based on load forecast data and grid stability requirements, determine the maximum and minimum loads in the selected time period, and calculate the peak-to-valley difference that the energy storage device needs to adjust; the technical evaluation includes comparisons of charging and discharging efficiency, power density, cycle life, cost and other indicators, and selects phase converters as the energy storage technology.

[0060] The calculation formula of the peak-to-valley difference is: ΔB = B max -B min ;

[0061] In the formula, ΔB is the peak-to-valley difference, that is, the power difference that needs to be adjusted by the energy storage device; B max is the maximum load in the selected time period, B min is the minimum load during the selected time period.

[0062] S42, calculating the energy difference within the selected time period according to the renewable energy power generation forecast data and the load demand forecast data within the selected time period;

[0063] The energy difference is calculated as follows: ΔC(t) = C lod (t)-C res (t);

[0064] Where ΔC(t) is the energy difference within time t, C lod (t) is the predicted amount of renewable energy power generation within time t, C res (t) is the load demand forecast within time t;

[0065] If the calculated energy difference is positive, it indicates an energy deficit, and if the calculated energy difference is negative, it indicates an energy surplus.

[0066] S43, determining the total energy storage capacity required by the energy storage device based on the energy difference;

[0067] The total energy storage is calculated as: D 总 =∑ t |ΔC(t)|;

[0068] Where D 总 is the total energy storage required by the energy storage device, and |ΔC(t)| is the absolute value of the energy difference within time t.

[0069] S44. In combination with the grid architecture and load distribution, energy storage systems are centrally deployed in substations and distribution stations for centralized control and scheduling.

[0070] Among them, a storage system management platform is established to realize remote monitoring, data analysis, fault diagnosis, intelligent scheduling and other functions; according to the needs of the power grid and the characteristics of the energy storage system, a reasonable energy management strategy is formulated to maximize utilization and save costs; finally, through repeated testing and adjustment, the safe and reliable operation of the energy storage system is ensured to achieve the expected performance indicators.

[0071] S5. Use intelligent control algorithms to optimize and supervise the grid-source load-storage system in real time and obtain real-time data; wherein the intelligent control algorithm selects RL algorithm for control, specifically,

[0072] Through RL algorithms and IoT technology, the status of the grid-source-load-storage system can be monitored in real time to detect potential faults and anomalies in a timely manner.

[0073] The calculation formula of the RL algorithm is:

[0074]

[0075] In the formula, E(s t ,q t ) is represented as state s t Next take action q t The expected future reward, s t It is represented as the grid state at time t, q t Indicates that in state s t Action taken under t Expressed as action q t and status t The rewards obtained are to encourage goals such as reducing grid operation costs and improving the efficiency of energy storage equipment; γ0 is the learning rate, which controls the speed of learning; δ is the discount factor, which represents the current value of future rewards. Represented as the next state s t+1 The maximum E value of all possible actions.

[0076] S6. Combine real-time data and load forecast data to obtain the optimal resource configuration optimization plan for the grid source, load and storage system, and dispatch grid resources according to the obtained optimal resource configuration optimization plan. Specifically,

[0077] Based on real-time data and load forecast data, conduct cost-benefit evaluations on different resource allocation plans and select the most economical plan; through resource allocation optimization and reasonable scheduling, reduce dependence on traditional energy, reduce electricity production costs, and minimize electricity operating costs; continuously monitor real-time data, update forecast model parameters, and continuously optimize resource allocation plans to achieve continuous minimization of grid operating costs.

[0078] Furthermore, based on the predicted output of renewable energy and the forecast of electricity demand, the output of renewable energy can be reasonably scheduled to achieve a balance between supply and demand. Then, based on the volatility of electricity demand and the volatility of renewable energy output, the energy storage equipment can be optimally configured to store excess electricity or release electricity under high load to improve energy utilization. Finally, based on electricity demand and renewable energy output, the operating strategies of traditional power generating units can be reasonably adjusted to reduce fuel costs and operating losses.

[0079] Embodiment 2, a network source load storage optimization device for a low voltage station area, comprising:

[0080] Data processing module: used to obtain the power grid resource data and historical load data of the low-voltage substation, investigate and evaluate the power grid resource data, and pre-process and extract features of the historical load data;

[0081] Model building module: used to build a load forecasting model based on the processed power grid resource data and historical load data, and use the historical load data to train the load forecasting model;

[0082] Load forecast data acquisition module: used to predict the load demand changes in the low-voltage area based on the trained load forecast model, and to obtain and monitor load forecast data;

[0083] Energy storage equipment deployment module: used to calculate the peak-to-valley difference and total energy storage capacity that the energy storage equipment needs to adjust according to load forecast data and grid stability requirements, and deploy energy storage equipment;

[0084] Real-time data acquisition module: used to optimize and monitor the grid-source, load-storage system in real time using intelligent control algorithms and acquire real-time data;

[0085] Optimization scheme design module: used to combine real-time data and load forecast data to obtain the optimal resource configuration optimization plan for the grid source, load and storage system, and perform grid resource dispatch based on the obtained optimal resource configuration optimization plan.

[0086] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for optimizing network source, load and storage in a low voltage area, characterized in that: include: Obtain the power grid resource data and historical load data of the low-voltage substation, investigate and evaluate the power grid resource data, and pre-process and extract features from the historical load data; Based on the processed power grid resource data and historical load data, a load forecasting model is constructed, and the load forecasting model is trained using the historical load data; Based on the trained load forecasting model, the load demand changes in the low-voltage area are predicted, and the load forecasting data is obtained and monitored; According to load forecast data and grid stability requirements, calculate the peak-to-valley difference and total energy storage capacity that the energy storage equipment needs to adjust, and deploy the energy storage equipment; Use intelligent control algorithms to optimize and monitor the grid-source, load-storage system in real time and obtain real-time data; Combine real-time data and load forecast data to obtain the optimal resource configuration optimization plan for the grid source, load and storage system, and perform grid resource dispatch based on the obtained optimal resource configuration optimization plan.

2. The method for optimizing the network source, load and storage in a low voltage area according to claim 1, characterized in that: The power grid resource data includes power supply capacity, power load, distribution network status, and energy structure.

3. The method for optimizing the network source, load and storage in a low voltage area according to claim 1, characterized in that: The historical load data includes load changes in different time periods and under different circumstances.

4. The method for optimizing the network source, load and storage in a low voltage area according to claim 1, characterized in that: The process of constructing a load forecasting model based on the processed power grid resource data and historical load data, and training the load forecasting model using the historical load data further includes: The cross-validation method is used to evaluate and optimize the load forecasting model.

5. The method for optimizing the network source, load and storage in a low voltage area according to claim 1, characterized in that: The peak-to-valley difference and the total energy storage amount that the energy storage device needs to adjust are calculated based on the load forecast data and the grid stability requirements, and the deployment of the energy storage device also includes: Based on load forecast data and grid stability requirements, conduct a technical evaluation of energy storage technology, determine the maximum and minimum loads in the selected time period, and calculate the peak-to-valley difference that the energy storage equipment needs to adjust; Calculate the energy difference within the selected time period based on the renewable energy generation forecast data and load demand forecast data within the selected time period; determining a total energy storage amount required by the energy storage device based on the energy difference; Combined with the grid architecture and load distribution, energy storage systems are centrally deployed in substations and distribution stations for centralized control and scheduling.

6. The method for optimizing the network source, load and storage of low-voltage substations according to claim 5, characterized in that: The calculation formula of the peak-to-valley difference is: ΔB = B max -B min ; In the formula, ΔB is the peak-to-valley difference, that is, the power difference that needs to be adjusted by the energy storage device; B max is the maximum load in the selected time period, B min is the minimum load during the selected time period.

7. The method for optimizing the network source, load and storage in a low voltage area according to claim 5, characterized in that: The energy difference is calculated as: ΔC(t) = C lod (t)-C res (t); Where ΔC(t) is the energy difference within time t, C lod (t) is the predicted amount of renewable energy power generation within time t, C res (t) is the load demand forecast within time t; The total energy storage is calculated as: D 总 =∑ t |ΔC(t)|; Where D 总 is the total energy storage required by the energy storage device, and |ΔC(t)| is the absolute value of the energy difference within time t.

8. The method for optimizing the network source, load and storage in a low voltage area according to claim 1, characterized in that: The use of intelligent control algorithms to optimize and supervise the grid-source load-storage system in real time and obtain real-time data also includes: Through RL algorithms and IoT technology, the status of the grid-source-load-storage system can be monitored in real time to detect potential faults and anomalies in a timely manner.

9. The method for optimizing the network source, load and storage in a low voltage area according to claim 1, characterized in that: The load forecasting model is calculated as: Y(t)=a0+a1X1(t)+……+a n X n (t)+α(t), Where Y(t) represents the predicted value of power load within time t, X i (t) is the observed value of the i-th influencing factor at time t, a i is the regression coefficient of the ith influencing factor, n is the total number of influencing factors, α(t) is the random interference term, α(t)~N(0, β 2 ), β 2 Expressed as the variance of the error term.

10. A device based on the network source load storage optimization method of the low voltage area according to claim 1, characterized in that: include, Data processing module: used to obtain the power grid resource data and historical load data of the low-voltage substation, investigate and evaluate the power grid resource data, and pre-process and extract features of the historical load data; Model building module: used to build a load forecasting model based on the processed power grid resource data and historical load data, and use the historical load data to train the load forecasting model; Load forecast data acquisition module: used to predict the load demand changes in the low-voltage area based on the trained load forecast model, and to obtain and monitor load forecast data; Energy storage equipment deployment module: used to calculate the peak-to-valley difference and total energy storage capacity that the energy storage equipment needs to adjust according to load forecast data and grid stability requirements, and deploy energy storage equipment; Real-time data acquisition module: used to optimize and monitor the grid-source, load-storage system in real time using intelligent control algorithms and acquire real-time data; Optimization scheme design module: used to combine real-time data and load forecast data to obtain the optimal resource configuration optimization plan for the grid source, load and storage system, and perform grid resource dispatch based on the obtained optimal resource configuration optimization plan.

Images