Power consumption behavior evaluation method for industrial park distributed power consumption network simulation analysis

Abstract

The invention relates to the technical field of power systems, and discloses an industrial park distributed power utilization network simulation analysis power utilization behavior evaluation method comprising the following steps: S1, data acquisition: power load data, equipment operation state data and environment data of each power utilization unit in an industrial park are acquired in real time, and the environment data comprise temperature and humidity; s2, data preprocessing: preprocessing the collected data to obtain complete data; and S3, simulation modeling: establishing a power load flow model based on the park information and the real-time data, and performing simulation modeling to generate a dynamic model under various power consumption situations. According to the invention, through real-time data acquisition, dynamic load scheduling and a deep reinforcement learning algorithm, flexible optimal distribution of electric power resources is realized, the efficiency and stability of park electric power management are improved, stable operation of equipment is ensured, and energy waste is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method for evaluating power consumption behavior through simulation analysis of a distributed power consumption network in an industrial park. Background Art

[0002] With the continuous expansion of industrial parks and the increase in the types of production equipment, the power demand in the park is becoming increasingly complex and diversified. In this context, how to efficiently and stably allocate and dispatch power resources, avoid energy waste and ensure the normal operation of equipment has become an urgent problem to be solved.

[0003] In the existing technology, many power management systems have begun to introduce optimization algorithms (such as linear programming, mixed integer programming, etc.) to dispatch the power load of the park. These systems can reasonably allocate electricity according to the power demand of the park, achieve a certain degree of energy saving effect, and ensure that power resources are not over-consumed. However, these systems usually rely on static load forecasts, and cannot adjust the power distribution plan in real time when facing fluctuations in power demand within the park. In addition, some systems use optimization algorithms to reduce the waste of power resources and ensure the stable operation of equipment, which has a certain resource optimization effect.

[0004] However, the existing technology still has some shortcomings and cannot fully cope with the complexity of modern campus electricity demand. First, most existing power dispatching methods rely on historical data or preset load models, which makes it impossible to adjust the dispatching plan in time according to the fluctuation of actual power demand, resulting in the power system being easily overloaded during peak power consumption or sudden load increase, and even equipment failure or unstable power supply. Secondly, the existing load optimization methods usually adopt a static dispatching mode, lack the ability to fully respond to real-time load data, and cannot flexibly respond to dynamic changes in equipment load or the influence of external environmental factors. Furthermore, although some advanced systems use optimization algorithms to dispatch electricity, most of them have not established a real-time feedback mechanism, resulting in the configuration of power resources often failing to achieve the best results. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a method for evaluating power consumption behavior through simulation analysis of a distributed power consumption network in an industrial park, which solves the problems in the prior art of being unable to respond to load fluctuations in real time and lacking flexible scheduling and optimization of power resource allocation.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for evaluating power consumption behavior of a distributed power consumption network in an industrial park by simulation analysis, comprising the following steps: S1. Data collection: Real-time collection of power load data, equipment operation status data and environmental data of each power-consuming unit in the industrial park, including temperature and humidity; S2, data preprocessing: preprocess the collected data to obtain complete data; S3, Simulation modeling: Based on the park information and real-time data, the power load flow model is established and simulated to generate dynamic models under various power consumption scenarios; S4. Electricity consumption behavior assessment: Conduct a behavior assessment on each electricity consumption unit in the park based on the simulation model to obtain information on electricity consumption patterns and load fluctuations; S5. Load dispatch optimization: load dispatch optimization is performed based on the evaluation results of power consumption behavior to reasonably allocate power resources; S6. Dynamic scheduling and feedback adjustment: Implement adaptive scheduling based on real-time data flow and evaluation results, and adjust the scheduling strategy based on feedback to ensure a stable supply of power resources.

[0007] Preferably, the data collection includes: Deploy smart meters and sensors at each power consumption unit in the industrial park to collect power load, current, voltage, power factor and equipment switch status in real time; The collected data is transmitted to a central control system using an industrial communication protocol.

[0008] Preferably, the data preprocessing includes: Time synchronization of collected data is performed to ensure the time consistency of each data source by using a timestamp alignment algorithm; De-noising the collected raw data, using Kalman filtering or mean filtering algorithms to remove outliers and noise; Normalize the data and convert the data of different dimensions into a unified scale range using the minimum-maximum standardization or Z-score standardization method to eliminate the impact of inconsistent dimensions on subsequent analysis; Interpolation is performed on missing data, and linear interpolation or spline interpolation algorithms are used to fill in the missing parts of the data to ensure data integrity.

[0009] Preferably, the simulation modeling includes: Establish a power flow model based on the load demand, equipment type and resource configuration of each power consumption unit in the park; Use time series analysis methods to model load fluctuations and generate load forecasting models; The simulation model is used to simulate the power demand in different working environments and form a prediction model for various power consumption scenarios.

[0010] Preferably, the electricity usage behavior assessment includes: Evaluate the electricity consumption behavior of each power consumption unit in the park based on the simulation model, and calculate the load fluctuation, peak demand and resource utilization efficiency of each unit; Output evaluation report, which includes information on load fluctuation, peak and valley values ​​and resource utilization efficiency of electricity consumption behavior.

[0011] Preferably, the load scheduling optimization includes: Based on the evaluation results of power consumption behavior, an optimization algorithm is used for load dispatch to ensure the rational allocation of power resources; The optimal load distribution strategy is adopted for different load units in the park to reduce power waste and avoid power overload.

[0012] Preferably, the dynamic scheduling and feedback adjustment includes: Use deep reinforcement learning algorithms to dynamically adjust power dispatch strategies based on real-time load data and power consumption behavior evaluation results; Adaptively adjust power dispatch through the feedback mechanism to optimize the use of power resources in the park.

[0013] Preferably, the configuration of the smart meter and the sensor includes: Install smart meters at the entrance of each power consumption unit in the park to monitor voltage, current, power factor and equipment status in real time; Temperature and humidity sensors are installed in key equipment and power consumption areas in the park to collect environmental data in real time.

[0014] Preferably, the power flow model includes: Establish power transmission relationships between power nodes in the park to form a power flow model that describes the transmission methods of current, voltage, and power; The power flow model is combined with the load forecasting model to conduct multi-dimensional simulation and evaluate electricity consumption behavior under different scenarios.

[0015] Preferably, the optimization algorithm includes: Linear programming is used to optimize the global dispatch of the park's power load; The mixed integer programming model is used to optimize the switching status of power equipment.

[0016] The present invention provides a method for evaluating power consumption behavior by simulating and analyzing distributed power consumption networks in industrial parks. It has the following beneficial effects: 1. The present invention collects data from each power-consuming unit in the park in real time and dynamically adjusts load scheduling in combination with a deep reinforcement learning algorithm to ensure the flexible allocation of power resources. Compared with the traditional static scheduling method, the present invention can respond quickly according to real-time changes, avoid power waste or overload operation, and ensure the efficient and stable operation of the system.

[0017] 2. The present invention uses optimization algorithms (linear programming and mixed integer programming) to accurately dispatch power loads and achieve optimal power distribution based on equipment characteristics and needs. Compared with the existing technology, the present invention can avoid uneven resource allocation, improve the overall efficiency of the power system, reduce unnecessary waste, and ensure accurate matching of the power demand of the park.

[0018] 3. The present invention ensures that power resources are adjusted in time according to equipment load fluctuations and external environmental changes through feedback mechanisms and adaptive adjustments. Compared with traditional methods, the present invention can flexibly deal with uncertainties, provide smarter load distribution, make park power management more efficient, avoid overload and waste of resources, and thus maximize energy utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] Please refer to the attached Figure 1 The embodiment of the present invention provides a method for evaluating power consumption behavior of a distributed power consumption network in an industrial park by simulation analysis, comprising the following steps: S1. Data collection: Real-time collection of power load data, equipment operation status data and environmental data of each power-consuming unit in the industrial park, including temperature and humidity; The core task of the data collection phase is to accurately and real-time obtain the power load, equipment operation status and environmental data (such as temperature and humidity) of each power unit in the park. These data will provide basic information for subsequent modeling, evaluation and optimization. Accurately collecting power data requires not only high-precision instruments and equipment, but also involves communication protocols, data transmission and other links, all of which affect the final evaluation results.

[0022] In this embodiment, the data collection method includes deploying smart meters and sensors, which are used to collect power load, current, voltage, power factor and device switch status respectively. Smart meters are the key to collecting power data. They can measure and record the current, voltage, power factor and other information of the power system in real time. Sensors are used to monitor environmental data, such as temperature and humidity, which have an important impact on the stability and efficiency of equipment operation.

[0023] In some embodiments, smart meters are installed at the entrance of each power unit, so that the power load flowing into the power unit can be accurately measured to ensure the comprehensiveness and accuracy of the data. For some key equipment or special areas, temperature and humidity sensors are installed between the equipment and power consumption areas in the park to monitor the environmental changes around the equipment in real time. Temperature and humidity data are crucial for evaluating the operating status of equipment and predicting power consumption patterns, especially in extreme weather or environmental changes, when the load of the equipment may fluctuate dramatically.

[0024] As an option, the data acquisition system uses industrial communication protocols (such as Modbus and OPC protocols) to transmit data to the central control system. These communication protocols can ensure that various meters and sensors can transmit data stably and efficiently in complex industrial environments, reduce delays and errors, and thus ensure the accuracy and real-time nature of the data. After the data is transmitted to the central control system through these protocols, it can be subsequently processed, analyzed, and simulated.

[0025] In some embodiments, in order to ensure that data collection is not interrupted, the system also adopts a redundant design. For example, multiple backup sensors and meters can immediately take over the data collection task when a failure occurs, ensuring the continuity and stability of the system. In this way, when a device fails, data collection can still be carried out normally, avoiding evaluation errors caused by data interruption.

[0026] In one possible implementation, the sampling frequency of smart meters and sensors can be dynamically adjusted according to the fluctuation of power consumption in the park. When the load fluctuates greatly, the system can automatically increase the sampling frequency to capture the subtle changes in the power load; when the load fluctuation is small, the system reduces the sampling frequency to optimize data storage and transmission efficiency. This flexible sampling strategy can better meet the needs of different power consumption units in the park.

[0027] Furthermore, the collected data includes not only power load data, but also the switch status of the equipment. The switch status of the equipment is a very important indicator in the evaluation of power consumption behavior, which reflects the working cycle and operation mode of the equipment. By monitoring the changes in the switch status of the equipment, the operation rules of the equipment in a specific time period can be identified, providing a reference for subsequent load scheduling.

[0028] In some embodiments, the power load is calculated as: ; in: It is electrical power, measured in watts (W), which is the power consumed or output by the device; is the voltage, measured in volts (V), which is the potential difference between the two ends of the device port; is the current, measured in amperes (A), which indicates the amount of charge flowing per unit time; It is the phase angle between current and voltage, expressed in degrees or radians, and reflects the relative position between the current and voltage waveforms; is the power factor, which is used to reflect the energy efficiency of the power system.

[0029] Voltage( ) is the potential difference generated when current passes through a conductor, which directly affects the power transmission efficiency. The unit of voltage is volt (V), which is usually provided by a power supply.

[0030] Current ( ) refers to the amount of charge flowing in a conductor, and its unit is ampere (A). It represents the amount of charge passing through the circuit per unit time, and the current determines the load of the electrical equipment.

[0031] Phase Angle ( ) is the phase difference between voltage and current, which is used to describe the degree of synchronization between the current and voltage waveforms. If there is a large phase difference in the equipment in the power system, it will usually affect the power factor, thereby reducing energy efficiency.

[0032] This formula is used to calculate the electric power of each power unit. According to the formula, electric power is directly related to voltage, current and power factor. In practical applications, power factor reflects the phase relationship between current and voltage and is usually used to measure the transmission efficiency of electricity. If the power factor is low, it means that the phase difference between current and voltage is large, which may lead to energy waste. Therefore, accurate measurement of voltage, current and power factor is crucial to improve the utilization efficiency of power resources.

[0033] In the present invention, smart meters and sensors use the above formula to calculate the power load in real time, providing basic data for subsequent simulation modeling and power consumption behavior evaluation. Whenever the voltage, current or power factor changes, the meter will recalculate the power according to the formula, thereby dynamically reflecting the load fluctuation of the power unit. These calculation results will be transmitted to the central control system in real time for subsequent analysis.

[0034] In this embodiment, the data acquisition module deploys smart meters and environmental sensors to collect power load, equipment status and environmental data in real time in each power consumption unit in the park. All collected data are transmitted to the central control system through the industrial communication protocol to ensure the accuracy and real-time nature of the data. In order to ensure the continuous and stable collection of data, the system adopts a redundant design and flexibly adjusts the sampling frequency to meet the needs of different power consumption scenarios. In the calculation of power load, the electric power calculation formula is used, which calculates the power load based on voltage, current and power factor, ensuring the comprehensiveness and accuracy of data collection. These data will provide reliable support for subsequent simulation modeling, power consumption behavior evaluation and load scheduling optimization.

[0035] S2, data preprocessing: preprocess the collected data to obtain complete data; Data preprocessing is a key step to ensure that the data collected from step S1 can be effectively used for subsequent analysis and modeling. Due to the complex working environment of multiple smart meters and sensors deployed in the industrial park, the collected data often have errors, inconsistencies and noise. The goal of data preprocessing is to ensure the accuracy of subsequent simulation modeling and electricity consumption behavior evaluation through a series of steps. Data preprocessing includes time synchronization, denoising, normalization and missing data interpolation of the collected data. The specific processing methods and technical implementation methods are as follows.

[0036] In an industrial park, there may be time differences between multiple meters and sensors, especially in distributed power systems, where sensors and meters may be located in different locations and not completely synchronized. In this embodiment, the time synchronization of various data sources is ensured by using a timestamp alignment algorithm. The timestamp alignment algorithm can automatically adjust the collected data so that data from different sensors can be accurately aligned to a unified time reference. Specifically, each data point is assigned a timestamp, and the system aligns the data by comparing these timestamps, thereby avoiding data deviations caused by time inconsistencies.

[0037] Generally speaking, the key to time synchronization is to eliminate the time differences between devices or sensors so that subsequent data analysis can be performed on the same time scale. By implementing this technology, the timing consistency of each data point can be ensured, providing a reliable foundation for subsequent data analysis and simulation modeling.

[0038] During the data collection process, the collected data may contain noise or outliers due to external interference or sensor errors. As an option, in order to eliminate the impact of these noises on subsequent analysis, Kalman filtering or mean filtering algorithms are used. Kalman filtering is a recursive filtering technique that can remove noise from dynamic data in real time and provide optimal smoothing based on state estimation. Mean filtering, on the other hand, eliminates local outliers by averaging the values ​​within the data window.

[0039] Specifically, for data with large real-time fluctuations, such as power load data, the advantage of Kalman filtering is that it can dynamically adapt to changes in system status and effectively remove high-frequency noise in time series. Mean filtering is suitable for scenarios with relatively stable data changes, which can reduce the interference of sudden outliers on data trends. The combination of the two can select the most appropriate denoising method according to the characteristics of different scenarios.

[0040] Since the data collected by different sensors have different dimensions (for example, the unit of current is ampere and the unit of temperature is Celsius), these differences will affect the accuracy and fairness of subsequent analysis. In this embodiment, in order to solve this problem, two normalization methods, minimum-maximum standardization and Z-score standardization, are used. Minimum-maximum standardization converts the data to a uniform scale range, usually between [0,1], thereby eliminating dimensional differences. Z-score standardization converts the data to zero mean and unit variance by calculating the mean and standard deviation of the data, so that the data can be compared under the same standard.

[0041] Specifically, the min-max normalization method is converted by the following formula: ; in: is the normalized data value; is the original data value; is the minimum value in the data set; is the maximum value in the data set.

[0042] As another option, Z-score normalization is handled by the following formula: ; in: is the standardized data value; is the original data value; is the mean of the data; is the standard deviation of the data.

[0043] These normalization methods ensure that all types of data can be processed at the same scale and avoid the impact of deviations between different dimensions on subsequent analysis.

[0044] During the data collection process, some data may be missing due to sensor failure or data loss. In some embodiments, in order to ensure the integrity and continuity of the data, a missing data interpolation method is used for processing. The missing data points can be estimated by linear interpolation or spline interpolation.

[0045] Linear interpolation: By using known adjacent data points, assuming that the missing data points are located in the linear relationship between these two points, an estimate is made. The specific formula is as follows: ; in: is the data value after interpolation; , is the value of the known data point; , is the timestamp of the known data point; The timestamp of the missing data.

[0046] Spline interpolation: Spline interpolation constructs a smooth curve to fit all points in the data, thereby obtaining an estimate of the missing points. Spline interpolation can provide more accurate interpolation results when the data changes more complexly, especially for scenarios with nonlinear changes.

[0047] As an option, if the missing data points are concentrated, more advanced interpolation algorithms can be used, such as polynomial interpolation or cubic interpolation. Different interpolation methods can be flexibly selected according to the characteristics of the data, thereby ensuring the integrity of the data and the accuracy of the analysis.

[0048] In this embodiment, the aforementioned electric power calculation formula is still used for the electric power load data; During the data preprocessing process, the values ​​of voltage, current and power factor may be abnormal or missing. Therefore, interpolation, denoising and normalization can ensure that these parameters can accurately reflect the actual situation.

[0049] Data preprocessing eliminates errors and inconsistencies in the data through technical means such as time synchronization, denoising, normalization and missing data interpolation, thereby providing a reliable data foundation for subsequent simulation modeling, power consumption behavior evaluation and load scheduling optimization.

[0050] S3, Simulation modeling: Based on the park information and real-time data, the power load flow model is established and simulated to generate dynamic models under various power consumption scenarios; Simulation modeling is to build a power flow model based on the collected pre-processed data and park information, and simulate the changes in power demand under different scenarios. The main goal of this step is to help evaluate the dynamic performance of the power system in the park by establishing an accurate power flow model and load fluctuation prediction model, thereby providing a scientific basis for subsequent load scheduling and resource optimization.

[0051] The power flow model is the basis of simulation modeling and is used to describe the transmission and distribution of power loads in the park. In this embodiment, we regard all power-consuming units in the park (such as production equipment, lighting equipment, air conditioning systems, etc.) as nodes in the power network and establish a power transmission relationship based on the power demand of each node to form a power flow model. The power flow model can fully reflect the distribution and flow of power resources in the park by calculating parameters such as current, voltage, and power between each node.

[0052] In general, the power flow model needs to consider the following parameters: current, voltage, power factor, etc. The current, voltage and power transmission relationship of each power node is solved by power flow algorithms (such as power flow calculation method). These calculations will simulate the power transmission path, transmission loss and voltage fluctuation, thus providing data support for the evaluation of the power demand of the park.

[0053] Specifically, the power flow model is usually calculated using network-based mathematical methods, such as DC power flow method, AC power flow method, etc. In this embodiment, the power flow model accurately simulates the power demand and current transmission path of each power consumption unit by considering factors such as load demand, equipment type, and resource configuration of each device in the park.

[0054] The power load in the park usually has obvious time fluctuations. As an option, this embodiment uses a time series analysis method to model load fluctuations. Load fluctuations usually follow certain periodic laws, especially in a specific time period or season, the load demand may fluctuate greatly. Therefore, using a time series analysis model (such as the autoregressive moving average model ARIMA, or the deep learning model LSTM) to predict load changes can provide reliable load demand forecasts.

[0055] Specifically, through time series analysis, future load demand can be predicted based on historical load data. This process generates a load forecast model for future time periods based on the patterns and trends of historical data. The output of the model can help predict the peaks, valleys, and fluctuations in power demand, thereby providing accurate data for the park's power resource scheduling and management.

[0056] In this embodiment, the power demand is not only affected by the load fluctuation of the power-consuming equipment, but also by different working environments and external factors. Specifically, the power demand of the park may fluctuate due to factors such as weather changes, holiday arrangements, and production plan adjustments. In order to comprehensively evaluate the dynamic changes in the power demand of the park, simulation modeling needs to simulate the power demand under multiple different scenarios.

[0057] In some embodiments, we consider the power consumption patterns in various scenarios, such as the difference in power demand between weekdays and weekends, the impact of weather changes on air conditioning and heating equipment, the periodic opening or closing of equipment, etc. Through these scenario simulations, more accurate load forecasts can be provided for the power dispatching of the park, helping managers to make resource allocation in advance.

[0058] In general, the power flow model and the load forecasting model need to be used in combination to fully evaluate the dynamic changes in the power demand of the park. The power flow model provides basic data of the power network, such as current, voltage and power transmission, while the load forecasting model predicts future load fluctuations through methods such as time series analysis. The combination of the two can provide a scientific basis for subsequent power consumption behavior evaluation and load scheduling optimization.

[0059] In one possible implementation, the combination of the power flow model and the load forecasting model can be calculated as follows: ; in: is the total electrical power, in watts (W), which represents the total power demand of all power-consuming units in the park; For the The voltage of each power unit, in volts (V); For the The current of each power unit, in amperes (A); For the The phase angle between the current and voltage of an electrical unit, in degrees (°) or radians (rad); is the number of electricity-consuming units in the park; For the The power factor of each power unit.

[0060] This formula is used to calculate the total power demand of all power consumption units in the park. By weighted calculation of the voltage, current and phase angle of each power consumption unit, the total power demand in the park is obtained. This calculation result can provide an accurate basis for subsequent load scheduling and resource optimization.

[0061] During the simulation modeling stage, by simulating the power flow within the park, key parameters such as the electric power, current, voltage and power factor of each power-consuming unit can be accurately calculated; and through timing analysis and scenario simulation, future power demand fluctuations can be accurately predicted. This process not only provides accurate data for subsequent power consumption behavior evaluation, but also provides a scientific basis for load scheduling optimization, helping the park to achieve more efficient power resource management.

[0062] S4. Electricity consumption behavior assessment: Conduct a behavior assessment on each electricity consumption unit in the park based on the simulation model to obtain information on electricity consumption patterns and load fluctuations; The power consumption behavior assessment is an important step to comprehensively evaluate the load fluctuation, peak demand and resource utilization efficiency of each power consumption unit in the park based on the results of previous data collection, data preprocessing, simulation modeling and other steps. Through this assessment, the power consumption pattern of each power consumption unit can be effectively revealed, helping park managers optimize load scheduling, improve energy utilization efficiency and avoid resource waste.

[0063] In this embodiment, the key indicators for evaluating power consumption behavior include load fluctuation, peak demand, and resource utilization efficiency. The calculation method of each indicator and its significance in park power management will be described in detail below.

[0064] Load fluctuation is an important indicator to measure the change of power demand of power consumption units. Generally, each device or power consumption unit in the park will have different power demand fluctuations according to factors such as working status, production cycle, equipment load, etc. As an option, the calculation method of load fluctuation is usually obtained by comparing the highest power and the lowest power in a certain period of time.

[0065] The calculation formula for load fluctuation is as follows: ; in: For the The load fluctuation of a power consumption unit indicates the variation of the power demand of the power consumption unit in a specific period of time; For the The maximum power of each power consumption unit during this time period, in watts (W); For the The minimum power of each power unit during this time period, in watts (W); No. The maximum electric power of an electric unit in a specific time period, which indicates the maximum power consumed by the electric unit during the peak load period; For the The minimum electric power of an electric unit in a specific time period indicates the minimum power consumed by the electric unit during the load valley period.

[0066] The larger the load fluctuation value, the more dramatic the change in the power demand of the power unit during the time period. By monitoring load fluctuations, the park management system can predict in advance which equipment or areas may cause power overload or shortage, and take appropriate load adjustment measures.

[0067] Peak demand refers to the maximum power demand of a power unit during peak load periods, which usually occurs during production peaks or specific working hours. In some embodiments, peak demand is considered an important indicator for evaluating whether a power unit needs to adjust its operating status. If the peak demand of a device is too high, it may cause an overload of the park power system and even affect the normal operation of other devices.

[0068] The peak demand is calculated as follows: ; in: For the The peak power of a power unit, which indicates the maximum power consumed by the device during a specific period of time, occurs during the peak load period, and is measured in watts (W); For the Power consumption units at time The electrical power at a certain moment, indicating the power consumption of the device at that moment, in watts (W); Indicates The maximum electrical power of an electrical unit in a certain period of time.

[0069] Peak demand can help managers identify which devices will generate larger power loads during specific periods of time, thereby giving priority to the power supply of these devices in actual power scheduling and avoiding uneven distribution of power resources due to excessive loads on other devices.

[0070] Resource utilization efficiency is an important indicator for evaluating the energy efficiency of equipment under different workloads. It indicates how efficiently the power-consuming unit utilizes electricity during fluctuations in power demand. Equipment with high resource utilization efficiency usually maintains relatively stable energy efficiency during load changes to avoid wasting electricity. Specifically, by comparing the average power and maximum power of a device, its energy efficiency performance under different load conditions can be measured.

[0071] The calculation formula for resource utilization efficiency is as follows: ; in: For the The resource utilization efficiency of each power unit, which indicates the energy efficiency of the equipment during its operation; For the The average power consumption of each power consumption unit, in watts (W); For the The maximum electrical power of a power unit, in watts (W).

[0072] The larger the value of resource utilization efficiency, the more efficient the power consumption of the device is, and it can maintain a good energy efficiency level under different workloads. In the power dispatch of the park, devices with higher resource utilization efficiency are usually given priority in the allocation of power resources to ensure the overall energy efficiency of the park.

[0073] Through comprehensive analysis of load fluctuations, peak demand, and resource utilization efficiency, the power consumption behavior of each power-consuming unit in the park can be comprehensively evaluated. In some embodiments, these evaluation results will be used to guide load scheduling and power resource allocation. For example, on devices with large load fluctuations, the system may prioritize load reduction to reduce the fluctuation of the overall power demand of the park; on devices with low resource utilization efficiency, the system may adopt optimization strategies to improve the operating efficiency of the equipment.

[0074] In this embodiment, the power consumption behavior assessment provides a comprehensive power demand analysis for the park by calculating load fluctuations, peak demand and resource utilization efficiency. Load fluctuations and peak demand can reveal the load changes of the park's power system, while resource utilization efficiency reflects the energy efficiency level of the equipment. Through these assessments, park managers can identify the utilization potential and potential optimization space of power resources, so as to carry out effective power dispatch and resource allocation, ensure the stable operation of the park's power system, and improve the overall energy utilization efficiency.

[0075] S5. Load dispatch optimization: load dispatch optimization is performed based on the evaluation results of power consumption behavior to reasonably allocate power resources; Through load scheduling optimization, power resources can be allocated scientifically based on the results of power consumption behavior evaluation, thereby improving the efficiency of the park's power system, avoiding power overload, and ensuring the stability and safety of power supply.

[0076] In this embodiment, load dispatch optimization is based on the load fluctuation, peak demand and resource utilization efficiency of each power unit obtained in the previous step S4 (power consumption behavior evaluation), and uses optimization algorithms (such as linear programming or mixed integer programming) to dispatch power loads to ensure efficient use of power resources. This process not only considers the allocation of power demand, but also combines the constraints of the power system, such as the maximum carrying capacity of the power system, equipment load restrictions, etc.

[0077] The main goal of load scheduling optimization is to evenly distribute the total power demand in the park to each power unit to avoid the situation where some equipment is overloaded or underloaded. Through load scheduling optimization, it is possible to reduce power waste as much as possible without overloading the power system and improve the overall power utilization efficiency of the park.

[0078] Generally, load dispatch optimization is solved by optimization algorithms (such as linear programming, mixed integer programming, etc.), which can minimize the consumption of power resources or maximize system efficiency while satisfying various constraints.

[0079] The goal of load dispatch optimization is usually to minimize the total power cost within the park or to improve the efficiency of the power system. The optimization objective function can be set based on power demand and cost, for example: ; in: For the The cost of electricity per unit of electricity consumption unit, in Yuan / W; For the The electrical power of each power unit, in watts (W); It is the total number of electricity-consuming units in the park.

[0080] The constraints in load scheduling optimization ensure the feasibility and safety of the scheduling scheme in actual operation. Common constraints include: Maximum carrying capacity of the power system: ; in: is the maximum carrying capacity of the park's power system, in watts (W); By calculating the total electric power in real time, the system can ensure that the total power demand does not exceed the maximum load of the system, thus avoiding system failure due to overload.

[0081] In one possible implementation, the power dispatching scheme needs to be adjusted dynamically as the load demand in the park changes. For example, as the workday ends or due to climate change, the load demand of some equipment in the park will fluctuate dramatically. At this time, the dispatching optimization algorithm can adjust the allocation of power resources in real time to maintain the stability of the system.

[0082] The key to dynamic scheduling is that the system can provide feedback and adjustments based on real-time data. By collecting the power demand and operating status of the equipment in real time, the scheduling system can respond to load fluctuations and adjust the power supply of the equipment to avoid resource waste or equipment overload. For example, when the system detects that a certain power unit consumes too much power during a high-load period, the system can optimize resource utilization by adjusting the unit's operating mode or allocating power to other low-load devices.

[0083] In some embodiments, the feedback mechanism plays a vital role in the load scheduling optimization process. By monitoring the load fluctuations in the park in real time, the system can dynamically adjust the scheduling strategy. For example, when a power unit is overloaded or has equipment failure, the system can immediately adjust the load distribution through the feedback mechanism to ensure the stability of the park's power supply.

[0084] The feedback mechanism can continuously optimize the power dispatch strategy through control algorithms (such as deep reinforcement learning or other adaptive algorithms). These algorithms continuously learn and adaptively adjust strategies based on real-time load data and evaluation results to ensure that the system is always in the optimal working state.

[0085] Load dispatch optimization uses an optimization algorithm to dynamically dispatch power resources within the park by comprehensively evaluating the results of power consumption behavior. This process not only takes into account the distribution of power demand, but also combines constraints and real-time load data to ensure the stability of power supply and efficient use of resources. By optimizing load dispatch, the park can maximize the operating efficiency of the power system, reduce power waste, ensure the normal operation of equipment, and avoid power overload while meeting power demand.

[0086] S6. Dynamic scheduling and feedback adjustment: Implement adaptive scheduling based on real-time data flow and evaluation results, and adjust the scheduling strategy based on feedback to ensure a stable supply of power resources; Dynamic dispatch and feedback adjustment are key steps to ensure that the power system can flexibly adjust according to real-time load changes and external environmental changes. Through dynamic dispatch and feedback mechanisms, the system can optimize the allocation of power resources, improve the stability and energy efficiency of the park power system, and avoid power waste or overload. This step is based on the previous load dispatch optimization results. With the support of real-time monitoring and feedback mechanisms, power distribution is dynamically adjusted to achieve optimal energy use.

[0087] In this embodiment, dynamic scheduling and feedback adjustment rely on the preliminary scheduling results generated by the previous load scheduling optimization step S5, and adjust the power distribution strategy through real-time data feedback to ensure the optimal configuration of power resources in the park. The system collects real-time data through the feedback mechanism and uses this data to adjust the load distribution, thereby optimizing the overall power efficiency of the park and the stability of the system.

[0088] The core goal of dynamic scheduling is to make flexible adjustments based on real-time load fluctuations in the park, equipment operating status, and external factors (such as climate, production demand, etc.) to ensure the rational allocation of power resources in the park. In general, dynamic scheduling needs to respond quickly to changes in power demand in the park while maintaining stable operation of equipment in the park. The scheduling system should optimize power distribution in real time based on a feedback mechanism to avoid waste of power resources and prevent overload or power shortage.

[0089] Specifically, dynamic scheduling collects power demand data of power consumption units in the park in real time and calculates the power required by each device in the next time period based on this data. The scheduling system performs power scheduling according to load fluctuations and adjusts the power distribution plan in real time to achieve optimal power utilization.

[0090] Feedback Mechanism and Power Demand Adjustment The feedback mechanism is a crucial part of dynamic scheduling. It monitors the actual power demand of equipment in the park and transmits data to the scheduling system in real time to ensure the optimal allocation of power resources. The system adjusts the power supply based on the deviation between real-time data and target demand. As an option, if the system finds that the load of certain power units deviates significantly from the forecast, the feedback mechanism will immediately transmit this information to the scheduling system for adjustment.

[0091] For example, suppose the difference between the actual power demand and the target power demand of a power unit is , you can use the following formula to adjust: ; in: For the Power consumption units at time The power demand adjustment at the moment, in watts (W); For the The target power demand of each power consumption unit, in watts (W), is usually pre-set in the load dispatch optimization step; For the Power consumption units at time The actual power demand at the moment, in watts (W).

[0092] In some embodiments, in order to enhance the adaptive ability of dynamic scheduling, the system uses self-learning algorithms such as deep reinforcement learning (DRL). Deep reinforcement learning continuously learns the optimal scheduling strategy through interaction with the environment, and is particularly suitable for complex and dynamic power demand scheduling. The algorithm can optimize the power scheduling strategy based on historical power consumption data and real-time feedback data to ensure the optimality of power resource allocation.

[0093] Specifically, the deep reinforcement learning algorithm evaluates the current power dispatch decision based on real-time load data, equipment operating status, external environmental changes and other factors, and adjusts the power distribution plan based on the feedback mechanism. The system guides the learning process through a reward mechanism (such as reducing power waste and improving resource utilization), so that the dispatch strategy can be continuously optimized in the long term.

[0094] The optimization objective can be expressed in the following form: ; in: For in time The dispatch reward at the moment, in dimensionless units. The reward reflects the optimization effect of the current dispatch decision. The larger the reward value, the more optimized the configuration of power resources; For the The cost of electricity per unit of electricity consumption unit, in Yuan / W; For the The power demand of each power consumption unit in watts (W); To adjust the penalty factor, the relationship between the system optimization goal and the real-time adjustment is weighed, usually a weight coefficient; It is the load adjustment amount, which reflects the adjustment made by the dispatching system based on real-time data.

[0095] By maximizing the scheduling reward ,Deep reinforcement learning algorithms can continuously optimize power dispatching decisions and ensure the efficient operation and stability of the power system.

[0096] Dynamic scheduling and feedback adjustment ensure the optimal allocation of power resources in the park through adaptive scheduling algorithms such as real-time monitoring, feedback mechanism and deep reinforcement learning. The system adjusts the power load through real-time data, optimizes the power supply of various equipment in the park, and avoids power waste and overload operation. The combination of feedback mechanism and deep reinforcement learning enables the system to flexibly respond to changes in power demand in the park, ensure the stability and efficiency of the power system, and continuously optimize the efficiency of the use of power resources in the park in long-term operation.

[0097] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The power consumption behavior evaluation method of the distributed power consumption network simulation analysis in the industrial park is characterized by: The following steps are involved: S1. Data collection: Real-time collection of power load data, equipment operation status data and environmental data of each power-consuming unit in the industrial park, including temperature and humidity; S2, data preprocessing: preprocess the collected data to obtain complete data; S3, Simulation modeling: Based on the park information and real-time data, the power load flow model is established and simulated to generate dynamic models under various power consumption scenarios; S4. Electricity consumption behavior assessment: Conduct a behavior assessment on each electricity consumption unit in the park based on the simulation model to obtain information on electricity consumption patterns and load fluctuations; S5. Load dispatch optimization: load dispatch optimization is performed based on the evaluation results of power consumption behavior to reasonably allocate power resources; S6. Dynamic scheduling and feedback adjustment: Implement adaptive scheduling based on real-time data flow and evaluation results, and adjust the scheduling strategy based on feedback to ensure a stable supply of power resources.

2. The method for evaluating power consumption behavior of distributed power consumption network in industrial parks according to claim 1 is characterized in that: The data collection includes: Deploy smart meters and sensors at each power consumption unit in the industrial park to collect power load, current, voltage, power factor and equipment switch status in real time; The data collection is transmitted to a central control system using industrial communication protocols.

3. The method for evaluating power consumption behavior of distributed power consumption network in industrial parks according to claim 1 is characterized in that: The data preprocessing includes: Time synchronization of collected data is performed to ensure the time consistency of each data source by using a timestamp alignment algorithm; De-noising the collected raw data, using Kalman filtering or mean filtering algorithms to remove outliers and noise; Normalize the data and convert the data of different dimensions into a unified scale range using the minimum-maximum standardization or Z-score standardization method to eliminate the impact of inconsistent dimensions on subsequent analysis; Interpolation is performed on missing data, and linear interpolation or spline interpolation algorithms are used to fill in the missing parts of the data to ensure data integrity.

4. The method for evaluating power consumption behavior of distributed power consumption network in industrial parks according to claim 1 is characterized in that: The simulation modeling includes: Establish a power flow model based on the load demand, equipment type and resource configuration of each power consumption unit in the park; Use time series analysis methods to model load fluctuations and generate load forecasting models; The simulation model is used to simulate the power demand in different working environments and form a prediction model for various power consumption scenarios.

5. The method for evaluating power consumption behavior of distributed power consumption network simulation analysis in industrial parks according to claim 1 is characterized in that: The electricity usage behavior assessment includes: Evaluate the electricity consumption behavior of each power consumption unit in the park based on the simulation model, and calculate the load fluctuation, peak demand and resource utilization efficiency of each unit; Output evaluation report, which includes information on load fluctuation, peak and valley values ​​and resource utilization efficiency of electricity consumption behavior.

6. The method for evaluating power consumption behavior of distributed power consumption network in industrial parks according to claim 1 is characterized in that: The load dispatch optimization includes: Based on the evaluation results of power consumption behavior, an optimization algorithm is used for load dispatch to ensure the rational allocation of power resources; The optimal load distribution strategy is adopted for different load units in the park to reduce power waste and avoid power overload.

7. The method for evaluating power consumption behavior of distributed power consumption network in industrial parks according to claim 1 is characterized in that: The dynamic scheduling and feedback adjustment include: Use deep reinforcement learning algorithms to dynamically adjust power dispatch strategies based on real-time load data and power consumption behavior evaluation results; Adaptively adjust power dispatch through the feedback mechanism to optimize the use of power resources in the park.

8. The method for evaluating power consumption behavior by simulation analysis of distributed power consumption networks in industrial parks according to claim 2, characterized in that: The configuration of the smart meter and sensor includes: Install smart meters at the entrance of each power consumption unit in the park to monitor voltage, current, power factor and equipment status in real time; Temperature and humidity sensors are installed in key equipment and power consumption areas in the park to collect environmental data in real time.

9. The method for evaluating power consumption behavior by simulation analysis of distributed power consumption networks in industrial parks according to claim 4, characterized in that: The power flow model includes: Establish power transmission relationships between power nodes in the park to form a power flow model that describes the transmission methods of current, voltage, and power; The power flow model is combined with the load forecasting model to conduct multi-dimensional simulation and evaluate electricity consumption behavior under different scenarios.

10. The method for evaluating power consumption behavior by simulation analysis of distributed power consumption network in industrial parks according to claim 6, characterized in that: The optimization algorithm includes: Linear programming is used to optimize the global dispatch of the park's power load; The mixed integer programming model is used to optimize the switching status of power equipment.

Images